{"id":"433ca6fc-edde-4969-b461-1afe0bcbcaca","arxiv_id":"2412.03322","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Two linearly uncoupled silicon nitride microrings sharing a short all-optically poled section achieve reconfigurable milliwatt-level second-harmonic generation over 1530-1620 nm and upconvert a 100 nm-wide Kerr comb.","lead":"The paper demonstrates a chip-scale device that frequency-doubles laser light across the telecom C and L bands with milliwatt output power, using two separately tuned silicon nitride rings that share a short nonlinear section. It matters because this allows broadband, reconfigurable, on-chip frequency conversion, useful for comb self-referencing and integrated nonlinear light sources.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The enabling premise of linearly uncoupled resonators is only locally established; the paper's own supplementary data show residual coupling and deliberate coupling for the MI comb, so independent addressing across the claimed band is not demonstrated.","rationale":"The reader's weakest assumption identifies the same load-bearing point: the device architecture works only if the two resonators remain effectively uncoupled while being independently tuned, and the paper's own supplementary characterization shows residual coupling that is only partially corrected by the MZI heater. I agree with the conditional verdict: the measured SHG powers, quadratic scaling, TPM-imaged grating, comb spectra, and the device's reconfigurability provide substantial independent support for the main demonstrations, so the concern does not warrant rejection. However, because the uncoupling premise enters the theoretical model and the claimed advantage over coupled-resonator approaches, and because the paper explicitly relies on the residual coupling to generate the broadband MI comb, the central claim requires a clearer statement of the parameter range over which linear uncoupling holds. The proposed test would settle this by quantifying the actual mode hybridization. The additional issue of the 10 mW and 100 nm metrics not being co-attained is present and should be fixed in presentation, but it is secondary to the uncoupling assumption and does not change the overall conditional assessment.","tokens_in":19720,"tokens_out":10370,"duration_ms":112990,"concrete_test":"On the device used for Fig. 5, record a low-power two-dimensional transmission map of the pump resonances versus north-heater power over one full SH FSR, with the MZI heater reset for each pump band. Fit the resonance trajectories to a two-coupled-mode eigenvalue problem and extract the residual coupling rate and supermode splitting; if the splitting exceeds the sum of the two linewidths, or if the pump resonance shifts by more than one linewidth when the SH resonance is tuned across the FSR, the linear-uncoupling assumption fails at that operating point and the independent-addressing claim must be restricted to configurations where the splitting is below linewidth.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on the two racetrack resonators being linearly uncoupled, so that pump and SH resonances can be addressed independently. This assumption enters the model directly: Methods Eqs. (2)-(3) set sigma_SH = kappa_FH = 1 and factorize the response into independent north and south field enhancements, while Eq. (5) treats each resonator as an isolated cavity. The manuscript's own evidence shows this premise is only approximate. Supplementary Note 1 reports residual linear coupling in the pump band, causing visibility loss, linewidth broadening, and an anticrossing-type hybridization with a north-resonator TE00 mode; the MZI heater restores near-ideal visibility only within a limited spectral region. Supplementary Note 4 then shows that for the modulation-instability comb the authors deliberately increase this coupling, producing an anticrossing-type perturbation of the integrated dispersion and linewidth broadening of higher-frequency resonances. Thus across the roughly 100 nm comb band the two resonators are demonstrably not uncoupled, and the MZI cannot null the coupling at all wavelengths simultaneously. If the residual coupling is significant at an operating point, the pump resonance is partly a supermode, the independent field-enhancement model in Eq. (5) is inaccurate, and the claimed advantage of avoiding supermode formation and dispersion alteration is not established. The independent-addressing paradigm therefore rests on a premise the manuscript itself shows to be locally, not globally, valid.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a silicon nitride photonic device consisting of two racetrack resonators—one for the pump (south) and one for the second harmonic (north)—that are intended to be linearly uncoupled but share a common interaction region where a photoinduced χ(2) grating is inscribed. The authors demonstrate reconfigurable doubly resonant second-harmonic generation across the C and L telecom bands with milliwatt-level output powers, and they use the same device to generate and upconvert a Kerr frequency comb with a pump-bandwidth of roughly 100 nm and upconverted power up to 10 mW. The central claim is that separating the pump and second-harmonic resonances into two independently addressable, linearly uncoupled resonators overcomes the bandwidth and tunability limitations of single-resonator SHG.","tokens_in":19915,"tokens_out":6783,"duration_ms":63994,"significance":"If the claims hold, this work is significant: it addresses a long-standing limitation of microresonator-based SHG—the simultaneous satisfaction of phase matching, double resonance, and FSR matching—by introducing a design in which the pump and SH resonances can be tuned independently. The experimental dataset is extensive and internally consistent in many respects: transmission spectra with loaded Q factors, SHG maps, a quadratic power-scaling fit with slope 2.02, two-photon microscope imaging of the χ(2) grating, and comb generation/upconversion spectra. The approach is CMOS-compatible and, in principle, transferable to other material platforms. However, the enabling premise of linear uncoupling is only partially verified, and several reported metrics are mutually inconsistent, so the strength of the central claim is not yet fully established.","major_comments":[{"comment":"The theoretical model sets σ_SH = κ_FH = 1, i.e., it assumes perfect linear uncoupling and factorizes the nonlinear response into independent north and south field enhancements. However, Supplementary Note 1 documents residual linear coupling in the pump band (visibility loss, linewidth broadening, and an anticrossing-type hybridization with a north-resonator TE00 mode) that the MZI heater can only compensate over a limited spectral window. Supplementary Note 4 then shows that for the MI comb the coupling is deliberately increased, producing an anticrossing-type perturbation of the integrated dispersion. Consequently, the assumption underlying Eq. (3) and Eq. (5) is not satisfied over the claimed operating bandwidth, and the factorization into independent field enhancements is an approximation whose accuracy is not quantified. The authors should quantify the residual coupling (e.g., from visibility and linewidth changes) and assess how it affects the predicted conversion efficiency and the independence of the two resonance families.","section":"Methods, Eq. (3); Supplementary Notes 1 and 4"},{"comment":"The Discussion states that a key advantage of the design is 'avoiding (or controlling) alterations to the dispersion profile associated with mode anti-crossings', but Supplementary Note 4 explains that the MI comb is formed by deliberately increasing the linear coupling, which 'produces an anticrossing-type perturbation ... facilitating the formation of MI combs'. These statements are in tension: the comb demonstration relies on a coupling that the paradigm is claimed to avoid. The manuscript should clarify that the uncoupled regime is an operating point used for SHG, whereas the coupled regime is intentionally exploited for comb generation, and that the device does not universally eliminate linear coupling.","section":"Discussion, first paragraph; Supplementary Note 4"},{"comment":"The paper reports a maximum SH power of 10 mW for a pump power of 220 mW and a maximum conversion efficiency of 40%/W. With CE = P_SH/P_FH^2, a CE of 40%/W at P_FH = 220 mW would yield P_SH ≈ 19 mW, roughly twice the reported maximum; conversely, 10 mW at 220 mW gives CE ≈ 20%/W. The authors should specify the pump conditions under which each value was obtained and explain whether thermal shifts, AOP reconfiguration, or other effects cause the CE to decrease at high power.","section":"Results, 'Addressable doubly resonant SHG'; Discussion"},{"comment":"The abstract claims 'milliwatt-level addressable second-harmonic generation over the entire telecom band', but the demonstrated range is 1530-1620 nm (the C and L bands, about 90 nm), and the text notes that this is limited by amplifier availability. The term 'entire telecom band' is an overstatement; the authors should either quantify the demonstrated range precisely or soften the claim (e.g., to 'C and L bands').","section":"Abstract; Fig. 2c"}],"minor_comments":[{"comment":"The notation for the coupling coefficients is confusing when σ_SH = κ_FH = 1; please clarify the values of σ_FH and κ_SH used for the numerical estimate.","section":"Methods, Eq. (3)"},{"comment":"The statement that TPM imaging 'confirm[s] the linearly uncoupled nature' is not direct evidence: TPM reveals the spatial distribution of χ(2), not the absence of linear coupling. Please rephrase to avoid over-interpretation.","section":"Fig. 2d-e"},{"comment":"Reference [26] appears to contain an arXiv identifier rather than a completed journal citation; please update it.","section":"References"},{"comment":"The conversion efficiencies across the C/L bands show considerable scatter (from about 1% to about 15%/W); the manuscript should comment on this wavelength dependence and its relation to the 'addressable' claim.","section":"Supplementary Figure 6"}],"recommendation":"major_revision","confidential_remarks":"The experimental work is substantial and likely publishable after revision, but the central 'linearly uncoupled' paradigm is not cleanly supported by the paper's own supplementary data, and the numerical inconsistency between the reported maximum CE (40%/W) and maximum SH power (10 mW at 220 mW) needs to be resolved. The tension between the Discussion's claim of avoiding dispersion alterations and Supplementary Note 4's use of deliberate coupling to generate the MI comb should be addressed head-on, as it bears on the novelty of the approach."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Plainly: this is a real device demonstration and the most convincing use of the linearly-uncoupled-resonator idea to date. The combination with all-optical poling for broad QPM, plus the synthetic FSR matching, gives a new architecture that delivers milliwatt-level SHG across the C+L band from a CMOS-compatible platform. The experimental record is strong: consistent transmission data, Q-factors near 0.7M/1.8M, SHG maps, a clean quadratic slope of 2.02, TPM imaging of the inscribed grating, and comb upconversion spectra. I believe the central claim—that the architecture solves the phase-matching/double-resonance/FSR trilemma better than a single ring—is supported.\n\nThe main caveats are about how the results are packaged and the limits of the uncoupled assumption. The abstract joins two records that are not achieved in the same configuration: the 10 mW per line and the 100 nm bandwidth come from different heater settings. That should be stated explicitly to avoid overreading. Also, the 'linearly uncoupled' premise is, by the paper's own admission, only locally exact. Supplementary Note 1 shows residual coupling broadening and an anticrossing feature in the pump band, corrected only in a limited spectral window; Supplementary Note 4 shows the MI comb generation actually uses deliberate coupling. This doesn't kill the SHG demonstration, because the device is reconfigured for each band, and the measured SHG doesn't rest on the model's assumption. But the theoretical estimate in Eq. (2)–(5) assumes perfect uncoupling, and the paper's claim about avoiding supermode formation should be qualified to the optimized operating points. The model gives CE ~22%/W vs ~40%/W measured—fine as a rough estimate, but the fit is loose.\n\nMinor: data/code are only 'available upon request'; for a record claim a public deposit would be better.\n\nWho should read it: anyone in integrated nonlinear photonics, microcombs, or frequency conversion. It deserves a serious referee; with a clarifying rewrite of the abstract and a frank supplementary statement about residual coupling, it's a clear accept. I'd bring it to the group.","headline":"A solid experimental advance in broadband on-chip SHG; the record numbers need unpacking and the 'uncoupled' premise is locally rather than globally true, but the core device works.","tokens_in":20540,"tokens_out":3077,"would_cite":true,"duration_ms":31123,"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":"Placing the pump and second-harmonic resonances in two linearly uncoupled microrings turns resonant frequency doubling into an ultrabroadband, reconfigurable process, with milliwatt output across the telecom band and upconverted Kerr…","keywords":["second-harmonic generation","microring resonators","silicon nitride photonics","all-optical poling","quasi-phase-matching","frequency comb upconversion","linearly uncoupled resonators","telecom-band nonlinear optics"],"falsifier":"Measure a single resonance pair while monitoring the bus waveguide on the north ring: if the rings are not effectively uncoupled, the pump resonance will split, broaden, or show an anticrossing at the point where the SH resonance crosses the pump, and the second-harmonic output will deviate from the product of the two independent Lorentzian field enhancements. A direct pass/fail experiment would be to set the MZI heater to the worst-case coupling, scan the north heater across the SH resonance, and check whether the SHG map loses its two independent branches.","tokens_in":19453,"feed_emoji":"🔬","tokens_out":8682,"duration_ms":75846,"temperature":0.7,"pith_summary":"This paper tries to establish that resonant frequency doubling does not have to be narrowband. The authors place the pump field and the second-harmonic field in two separate, linearly uncoupled racetrack resonators that share only a short interaction arm, so the resonance condition for each color can be set independently. A quasi-phase-matching grating written optically into the shared arm then allows frequency doubling over a bandwidth exceeding 200 nm. The device is reported to produce milliwatt-level second-harmonic output across the telecom C and L bands, and to upconvert an internally generated Kerr comb spanning about 100 nm with up to 10 mW per line. If correct, this removes the previous trade-off between conversion efficiency and spectral coverage for resonant frequency doubling.","feed_headline":"Two uncoupled rings double light across the whole telecom band","feed_subtitle":"Separate pump and harmonic resonators lift chip frequency doubling from one line to the whole telecom C and L bands.","key_machinery":"The central object is a pair of racetrack resonators joined by a Mach–Zehnder interferometer whose directional couplers are designed to split 100:0 in the pump band and 0:100 in the harmonic band, so the two rings share a physical path without sharing resonant modes. The mechanism that carries the argument is the all-optically poled quasi-phase-matching grating written in the shared arm, with period $\\Lambda\\approx 4.35\\,\\mu$m satisfying $\\Delta k=k_{\\mathrm{SH}}-2k_{\\mathrm{FH}}-2\\pi/\\Lambda=0$. Its bandwidth is set by the short grating length rather than by resonator dispersion, and the independent heaters on each ring plus the heater on the interferometer let the experimenter restore the doubly resonant condition and suppress residual linear coupling. This separation of functions is what allows many resonance pairs to participate in frequency doubling at once and what lets a Kerr comb be upconverted line by line.","core_discovery":"The central claim is that the three constraints that forced narrowband operation — phase matching, the doubly resonant condition, and free-spectral-range matching — can be treated separately by assigning pump and second harmonic to different resonators. The south ring is resonant at the pump, the north ring at the harmonic, and the only nonlinear contact between them is a Mach–Zehnder interaction arm in which both fields circulate together and all-optical poling writes a $\\chi^{(2)}$ grating. Because the rings are linearly uncoupled, the two resonance families can be shifted independently with on-chip heaters, and the loop lengths can be chosen so that $v_g^{\\mathrm{FH}}/L^{\\mathrm{FH}}=v_g^{\\mathrm{SH}}/L^{\\mathrm{SH}}$, yielding matched free spectral ranges without dispersion engineering. The paper reports addressable second-harmonic generation at milliwatt power from 1530 to 1620 nm, and then configures the same device to generate an incoherent modulation-instability comb and frequency-double it to an upconverted comb spanning nearly 50 nm in the harmonic band (about 100 nm in the pump band), with up to 10 mW per line.","pith_inferences":["Inference: the linearly uncoupled design could be modularized, with a pump cavity and harmonic cavity optimized separately on different materials and connected by a nonlinear waveguide, which would extend this architecture to wavelength bands where a single CMOS-compatible material lacks the needed resonances.","Inference: the observed pump-depletion trade-off between comb formation and upconversion suggests that assigning different lengths of the shared arm to $\\chi^{(2)}$ and $\\chi^{(3)}$ processes, or tuning their relative strengths, could allow broadband comb generation and efficient upconversion simultaneously rather than as a compromise.","Inference: because the grating is written and erased optically, the same device could be tested as a programmable frequency translator by sequentially poling different resonances and checking whether old gratings survive or must be rewritten, which would determine how quickly the device can be reconfigured in a real system."],"forward_implications":["The singly resonant microring trade-off between conversion efficiency and bandwidth is replaced by a design in which the two are set by separate degrees of freedom.","A chip-scale source can deliver milliwatt-level second-harmonic light at any addressable telecom resonance, with conversion efficiency up to 40%/W in continuous wave operation.","Frequency combs generated in the pump ring can be upconverted on the same chip, giving per-line powers above 10 mW over roughly 100 nm of pump bandwidth.","The electrically reconfigurable doubly resonant condition means fabrication tolerances no longer decide which resonance pair can be used, and a synchronized pump-and-heater scan could in principle give gap-free tuning of the harmonic wavelength."],"supporting_citations":[{"why":"introduced nonlinear coupling of linearly uncoupled resonators, the concept this device implements for pump and harmonic fields.","marker":"[30]"},{"why":"demonstrated all-optical poling in silicon nitride microresonators, providing the photoinduced chi(2) grating used here.","marker":"[20]"},{"why":"demonstrated photoinduced second-harmonic generation in silicon nitride microrings, the efficiency baseline this work broadens.","marker":"[19]"},{"why":"showed self-organized nonlinear gratings and was used for the two-photon microscopy imaging of the inscribed grating.","marker":"[17]"},{"why":"demonstrated broadband quasi-phase-matching in dispersion-engineered all-optically poled silicon nitride waveguides, the dispersion-engineering route this design avoids.","marker":"[21]"},{"why":"established formation rules and long-lived stability of photoinduced gratings, which supports the cold-cavity regime measurements.","marker":"[36]"},{"why":"explained the dynamics of self-organized spatiotemporal quasi-phase-matching used to interpret the AOP onset and the drop of the SHG signal.","marker":"[37]"},{"why":"provided the normal-dispersion Kerr comb framework used to interpret the modulation-instability comb and the anticrossing-assisted generation.","marker":"[38]"},{"why":"earlier single-resonator demonstration of near-visible comb upconversion with much lower power, the comparison for the upconverted comb performance.","marker":"[27]"},{"why":"periodically poled lithium niobate microring with high conversion efficiency, one of the benchmarks in the state-of-the-art table.","marker":"[9]"}],"fun_headline_variants":["Chip doubles light across entire telecom band with two rings","Milliwatt doubling on a chip now spans telecom C and L","Broadband frequency doubling via two independent microresonators","Integrated device doubles laser frequency over 200-nm range","Two-ring system lifts frequency doubling to wide bandwidth"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the two rings remain effectively linearly uncoupled across the operating band after tuning the Mach–Zehnder heater; the paper's own data show residual coupling as visibility loss, linewidth broadening, and an anticrossing perturbation, so if that coupling cannot be controlled, independent addressing of pump and harmonic resonances fails.","fun_headline_variants_meta":{"raw":{"variants":["Chip doubles light across entire telecom band with two rings","Milliwatt doubling on a chip now spans telecom C and L","Broadband frequency doubling via two independent microresonators","Integrated device doubles laser frequency over 200-nm range","Two-ring system lifts frequency doubling to wide bandwidth"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000459,"raw_usage":{"total_tokens":2334,"prompt_tokens":1015,"completion_tokens":1319,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":631,"completion_tokens_details":{"reasoning_tokens":1238}},"tokens_in":631,"tokens_out":1319,"duration_ms":12601,"temperature":1.0,"reasoning_tokens":1238,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:31:50.782197+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure a single resonance pair while monitoring the bus waveguide on the north ring: if the rings are not effectively uncoupled, the pump resonance will split, broaden, or show an anticrossing at the point where the SH resonance crosses the pump, and the second-harmonic output will deviate from the product of the two independent Lorentzian field enhancements. A direct pass/fail experiment would be to set the MZI heater to the worst-case coupling, scan the north heater across the SH resonance, and check whether the SHG map loses its two independent branches.","supporting_citations":[{"cited_title":"Menotti, B","cited_arxiv_id":null,"evidence_quote":"introduced nonlinear coupling of linearly uncoupled resonators, the concept this device implements for pump and harmonic fields."},{"cited_title":"Nitiss, J","cited_arxiv_id":null,"evidence_quote":"demonstrated all-optical poling in silicon nitride microresonators, providing the photoinduced chi(2) grating used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"demonstrated photoinduced second-harmonic generation in silicon nitride microrings, the efficiency baseline this work broadens."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"showed self-organized nonlinear gratings and was used for the two-photon microscopy imaging of the inscribed grating."},{"cited_title":"Nitiss, B","cited_arxiv_id":null,"evidence_quote":"demonstrated broadband quasi-phase-matching in dispersion-engineered all-optically poled silicon nitride waveguides, the dispersion-engineering route this design avoids."},{"cited_title":"Nitiss, T","cited_arxiv_id":null,"evidence_quote":"established formation rules and long-lived stability of photoinduced gratings, which supports the cold-cavity regime measurements."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"explained the dynamics of self-organized spatiotemporal quasi-phase-matching used to interpret the AOP onset and the drop of the SHG signal."},{"cited_title":"Guo, C.-L","cited_arxiv_id":null,"evidence_quote":"earlier single-resonator demonstration of near-visible comb upconversion with much lower power, the comparison for the upconverted comb performance."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"periodically poled lithium niobate microring with high conversion efficiency, one of the benchmarks in the state-of-the-art table."}],"review_version":1}