{"id":"8eb8a69c-0db3-4541-b034-e942092f309a","arxiv_id":"2412.10403","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The spectral hole linewidth in Eu:YSO increases linearly with temperature from 0.1 to 1 K, despite negligible expected T^7 Raman broadening.","lead":"Researchers measured ultranarrow spectral holes in a europium-doped crystal at temperatures from 0.1 to 1 kelvin and found the holes widen linearly as temperature rises, instead of staying constant as standard theory predicts. The result matters for building compact, ultra-stable laser frequency references that could replace bulky optical cavities.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The three slopes in Fig. 3 differ by up to a factor of two, and the paper does not model how burn/probe convolution converts homogeneous-linewidth temperature dependence into measured hole width; a single intrinsic linear slope is not yet established.","rationale":"I read the paper in good faith. The qualitative observations—linear broadening of spectral-hole linewidth with temperature below 1 K and a phase discriminant exceeding 0.6 mrad/Hz—are interesting and supported by several controls, including the reversibility of the heated hole and the 100 nW power threshold for visible spectral diffusion. However, the quantitative interpretation of the three linewidth slopes is the weakest link. A spectral hole width is not a direct readout of the homogeneous linewidth at the probe temperature; it is a convolution of the hole written at the burn temperature and the absorption line at the probe temperature. Thus Fig. 3(a) and Fig. 3(b) are expected to give different slopes even for a single intrinsic Gamma_hom(T). The close-to-factor-of-two ratio between 0.48 kHz/K and 0.95 kHz/K is exactly the pattern expected if the burn-time and probe-time homogeneous widths are additive, yet the paper dismisses the spread as 'procedures differ significantly' without analyzing this convolution. The reader's weakest assumption—that the measured hole width equals the true homogeneous linewidth—captures the same root issue, but the more specific missing piece is not residual power broadening or overburning alone; it is the unmodeled burn/probe convolution and the consistency of the three datasets. A global refit with separate burn and probe contributions would settle whether a single linear Gamma_hom(T) can describe all three protocols. If the refit succeeds, the linear broadening is robust; if it fails, the slopes are protocol-dependent and the title's claim about homogeneous linewidth behaviour is overreaching. Because the paper already includes relevant controls and the linear-T mechanism has precedent in the literature, I would not overturn the conditional acceptance; I would keep the reader's verdict and add the global-convolution analysis as an explicit condition. This is why I set verdict_should_be to UNCHANGED and agreement_with_reader to partial.","tokens_in":8111,"tokens_out":13667,"duration_ms":149833,"concrete_test":"Obtain the underlying linewidth data for Fig. 3 and re-fit all three datasets with a model that separates burn-time and probe-time homogeneous contributions, e.g. W_obs(T) = W0 + a*Gamma_hom(T_burn) + b*Gamma_hom(T_probe), with Gamma_hom(T) = Gamma0 + beta*T, using T_burn = 100 mK for Fig. 3(a) and T_burn = T_probe for Fig. 3(b). If a single beta cannot describe all three datasets within uncertainties, the protocol-dependent slopes are inconsistent with a common linear homogeneous-linewidth mechanism and the central claim weakens; if a single beta does describe them, the linear broadening is confirmed and the reported slopes are reconciled.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In a persistent spectral hole, the measured transmission profile is a convolution of the population hole written at the burn temperature and the homogeneous absorption line at the probe temperature. Therefore the slope of hole FWHM versus T is not simply the slope of the homogeneous linewidth: Fig. 3(a) burns at 100 mK and probes at each T, while Fig. 3(b) burns and probes at the same T. These two protocols should yield different slopes even for the same intrinsic Gamma_hom(T), and indeed the same-hole slope (0.48 kHz/K) is roughly half the burn-and-probe-at-T slopes (0.82 and 0.95 kHz/K). The paper notes that the prefactors differ but does not model this convolution, so it has not shown that the three datasets are mutually consistent or that the linear slopes reflect an intrinsic homogeneous-linewidth mechanism rather than a protocol-dependent artifact. The central quantitative claim—that Gamma_hom(T) increases linearly with T below 1 K—is therefore not yet pinned down.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports spectral-hole-burning measurements in Eu3+:Y2SiO5 at dilution-refrigerator temperatures (100 mK to 1 K), motivated by frequency-stabilization applications. The authors characterize hole linewidth and phase discriminant versus burn power and duration, obtaining a phase discriminant of (0.64 ± 0.06) mrad/Hz in the 100–300 mK range, more than twice their earlier value at 3.5 K. They then study the temperature dependence of the hole linewidth using three protocols: (a) a single hole burned at 100 mK and progressively heated and cooled, (b) independent holes burned and probed at each temperature, and (c) extrapolation of burn-duration scans to zero duration. In all three cases they report a linear increase of hole linewidth with temperature, with slopes of (0.48 ± 0.04), (0.82 ± 0.05), and (0.95 ± 0.06) kHz/K, despite the expected T^7 two-phonon Raman contribution being only about 0.04 Hz up to 1 K. The paper interprets this as evidence for an underlying linear broadening mechanism, possibly related to two-level systems, requiring further investigation.","tokens_in":8334,"tokens_out":3013,"duration_ms":34170,"significance":"If the central claim is robust, the observation of a linear (rather than T^7) linewidth increase in a nominally ordered crystal in the sub-kelvin regime is of genuine interest for solid-state emitter physics and for frequency-referencing applications. The reported phase-discriminant improvement, a factor of two over previous work at 3.5 K, is an experimentally concrete result with direct practical value, and the reversibility check on the heated hole is a useful control. However, the quantitative claim that the homogeneous linewidth itself grows linearly with temperature is not yet established, because the paper does not model how the measured hole profile relates to the homogeneous linewidth for the different burn/probe protocols, and because the three reported slopes differ by up to a factor of two without a demonstrated common origin.","major_comments":[{"comment":"The measured quantity is the spectral-hole FWHM, not the homogeneous linewidth directly. In persistent hole burning, the hole profile is a convolution of the population hole written at the burn temperature with the homogeneous absorption line at the probe temperature. For a hole burned at 100 mK and probed at temperature T (Fig. 3(a)), the slope of the measured FWHM versus T is not the slope of Gamma_hom(T); similarly, for holes burned and probed at the same T (Fig. 3(b), black circles), the convolution introduces a different conversion factor. The observed factor-of-two difference between the same-hole slope (0.48 kHz/K) and the burn-and-probe-at-T slopes (0.82 and 0.95 kHz/K) is qualitatively consistent with such a convolution effect. The paper states that 'we do not expect a quantitative agreement' but does not provide a model to show that the three datasets are mutually consistent with a single intrinsic Gamma_hom(T). This is a load-bearing omission: without it, the central claim that the homogeneous linewidth increases linearly with temperature below 1 K is not established.","section":"Fig. 3 and the paragraphs following it"},{"comment":"The third dataset is obtained by extrapolating the burn-duration scans in Fig. 2(c) to zero burn duration using ad hoc fits of the form a x^b + x0. The manuscript does not report the fitted parameters, the uncertainties of the extrapolated intercepts x0, or the goodness of fit. The extrapolation is used as a third measurement of the temperature-dependent linewidth, so its validity matters directly for the claimed linear dependence. I ask the authors to provide the fit parameters, residuals, and uncertainties, and to discuss whether the extracted x0(T) is robust against the choice of fitting function (for example, a pure power law without an intercept, or a saturating function derived from a rate-equation model of the burning process).","section":"Fig. 2(c) and Fig. 3(b), red squares"},{"comment":"Individual data points and their uncertainties are not tabulated for the linewidth-versus-temperature measurements. Given that the three slopes differ by roughly a factor of two, the reader cannot assess whether the differences are statistically significant or whether the linear fits are dominated by a few points. Please include a table (or supplementary material) with all burn parameters, measured FWHM values, uncertainties, and the linear-fit results (including intercepts and chi-squared). This is necessary to evaluate the consistency of the three protocols.","section":"Fig. 3 and Fig. 2(c)"},{"comment":"The paper states that no spectral-hole broadening is observed for optical powers of 100 nW and below, but this threshold is established at the temperatures shown in Fig. 2, and residual power broadening or instantaneous spectral diffusion that varies with temperature is not explicitly excluded. If such a temperature-dependent artifact contributes to the measured hole width, it could mimic a linear slope without reflecting the intrinsic homogeneous linewidth. I recommend reporting the power-dependence data at more than one temperature (or providing an estimate of the residual broadening from the Fig. 2(c) fits at each temperature) to show that the 10 nW/0.1 s burn parameters used for the black circles in Fig. 3(b) are in the power-independent regime at every temperature.","section":"Fig. 2 and the paragraph on instantaneous spectral diffusion"}],"minor_comments":[{"comment":"There are typographical errors: 'authours' in the acknowledgments should be 'authors', and 'dependant' should be 'dependent' in the paragraph discussing crystal fabrication. 'Raman' is capitalized inconsistently (e.g., 'two-phonon Raman' vs 'raman').","section":"General typography"},{"comment":"The caption states that the dotted lines represent 'the expected linewidth based on two-phonon raman broadening alone,' but the text mentions two values of alpha (0.0072 and 0.044 Hz/K^7) from different references. Please specify in the caption which alpha value was used to generate the dotted lines, or state that both values yield lines indistinguishable on the plot scale.","section":"Fig. 3 caption"},{"comment":"The black circles around data points in Fig. 2(a)–(d) are used to identify the specific burn durations shown in the top panels, but the figure caption does not define this marker convention. Please add a definition in the caption or legend.","section":"Fig. 2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript reports an interesting and potentially publishable measurement, but the central quantitative claim requires either a model of the hole-burning/probing convolution converting the three measured slopes into a homogeneous-linewidth temperature dependence, or a reformulation of the claim as a protocol-dependent hole-linewidth observation. The absence of tabulated data and fit details for Fig. 3 is a reproducibility concern that should be addressed regardless. I recommend major revision rather than rejection because the experimental results appear solid and the missing analysis is within the scope of a revised manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth a serious look. What's actually new: the first sub-kelvin Eu:YSO linewidth dataset, a phase discriminant of 0.64 mrad/Hz at 100–300 mK (about a two-fold improvement over their 3.5 K work), and three protocols that all show linear temperature broadening. The qualitative linear dependence was already reported in Eu:YSO at 1.5–5.5 K and in other systems, so the headline isn't a new mechanism—but extending to below 1 K, where the T^7 Raman contribution is predicted to be negligible, is a real experimental result.\n\nThe experiments look careful. They verify a 100 nW power threshold below which they see no spectral diffusion or power broadening, they show the same hole's linewidth returns to its initial value after a heating/cooling cycle (ruling out cumulative overburning), and they measure with three different procedures. The stress test's worry about the factor-of-two spread in fitted slopes does not actually hold up. Burn-and-probe at the same temperature measures a hole width that is roughly twice the homogeneous linewidth (convolution of the burn and probe Lorentzians), while burning at 100 mK and then probing at higher T measures a convolution of the narrow low-T hole with the probe line, whose slope should be the intrinsic homogeneous slope. So the observed slopes—0.48 kHz/K for same-hole, 0.82 and 0.95 kHz/K for burn-and-probe—are consistent with a single intrinsic slope of roughly 0.45–0.5 kHz/K, within the factor of two you'd expect. The paper is sloppy for not explicitly stating this, but the data are not inconsistent with a single intrinsic mechanism.\n\nSoft spots: no raw data or per-point uncertainties are tabulated; the zero-burn-duration extrapolation relies on an ad hoc power-law fit; and the authors could have strengthened the central claim by modeling the convolution instead of just saying the prefactors differ. None of these are fatal. The paper is honest about prior work and about the tentative TLS attribution, and it explicitly calls for specific-heat measurements as a follow-up.\n\nThis will be useful for people building spectral-hole frequency references and for anyone studying low-temperature dephasing in rare-earth-doped crystals. It deserves peer review. I'd send it to a serious referee, with the request that the authors add raw data and a short paragraph on the convolution factor.","headline":"Solid sub-kelvin Eu:YSO spectral-hole data with a useful metrology payoff; the factor-of-two slope spread is explained by standard burn/probe convolution and is not a flaw.","tokens_in":8940,"tokens_out":4349,"would_cite":false,"duration_ms":50205,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["42.50.Wk","42.50.Ct","76.30.Kg"],"model":"deepseek-v4-flash","headline":"Spectral holes in Eu:YSO broaden linearly with temperature from 100 mK to 1 K, even though two-phonon Raman theory predicts almost no broadening.","keywords":["spectral hole burning","homogeneous linewidth","Eu:YSO","sub-kelvin temperatures","two-level systems","phase discriminant","slow light","frequency stabilization"],"falsifier":"Measure the homogeneous linewidth of the same Eu:YSO crystal between 100 mK and 1 K with a two-pulse photon-echo sequence that does not rely on hole burning: if the echo linewidth stays flat while the hole width climbs linearly with temperature, the linear slope is an artifact of the burning or probing protocol. A complementary check is a specific-heat measurement in the same range, since an excess linear-in-$T$ term would independently support the two-level-system interpretation.","tokens_in":7956,"feed_emoji":"❄️","tokens_out":7982,"duration_ms":69984,"temperature":0.7,"pith_summary":"This paper reports that the ultra-narrow spectral holes burned in a Eu$^{3+}$:Y$_2$SiO$_5$ crystal do not keep a constant width as the crystal is cooled toward absolute zero. Between roughly 100 mK and 1 K, where two-phonon Raman theory predicts essentially no temperature broadening (about 0.04 Hz), the measured hole linewidth instead grows linearly with temperature at rates of 0.48 to 0.95 kHz/K. The authors read this as evidence that a mechanism other than Raman scattering, likely disorder modes or two-level systems, controls the homogeneous linewidth in this regime. They also show that the phase-discrimination signal useful for laser frequency locking more than doubles compared with their previous 3.5 K work, reaching $(0.64 \\pm 0.06)$ mrad/Hz.","feed_headline":"At 0.1–1 K, spectral holes widen linearly with heat","feed_subtitle":"Expected T^7 phonon broadening adds only 0.04 Hz; measured slopes reach ~1 kHz/K, hinting at two-level systems.","key_machinery":"The central object is the spectral hole burned into the inhomogeneously broadened $^7F_0 \\rightarrow ^5D_0$ transition of Eu$^{3+}$ ions in a Y$_2$SiO$_5$ host, probed with a frequency-controlled heterodyne laser. Its full width at half maximum is taken as the homogeneous linewidth of the emitters. The argument turns on comparing this measured width with the expected $T^7$ two-phonon Raman contribution, whose coefficient is small enough (up to 0.044 Hz/K$^7$) that the predicted increase from 0 to 1 K is only about 0.04 Hz; the observed linear slopes are therefore three to four orders of magnitude larger. A second key quantity is the phase discriminant, the slope of the optical phase shift versus frequency detuning at the hole centre, which determines how well the hole can serve as a frequency-locking reference.","core_discovery":"The central discovery is that the homogeneous linewidth of spectral holes in Eu:YSO follows a linear temperature dependence in the sub-kelvin range, contrary to the flat behaviour expected from the $T^7$ two-phonon Raman law. In three independent protocols, heating and cooling a single hole, burning fresh holes at each temperature, and extrapolating burn-duration scans to zero duration, the same linear trend appears, with slopes $(0.48 \\pm 0.04)$, $(0.82 \\pm 0.05)$, and $(0.95 \\pm 0.06)$ kHz/K. The authors argue that the persistence of a linear term is consistent with two-level-system or disorder-mode dephasing, and note that their crystal would not have been classified as anomalous on earlier criteria, suggesting these defects are a matter of degree rather than all-or-nothing. The same measurements show that burn power and duration must be optimised per temperature; at 100 nW power the best phase discriminant reaches $(0.64 \\pm 0.06)$ mrad/Hz between 100 and 300 mK, corresponding to a group delay of about 100 microseconds and an effective light speed in the 4 mm crystal near 40 m/s.","pith_inferences":["The authors leave implicit that their three slopes differ (0.48 vs 0.82 and 0.95 kHz/K) partly because the protocols heat or burn differently; a power series at fixed temperature could separate the intrinsic homogeneous term from any residual burn-induced broadening.","If two-level systems are responsible, the same defect population should show up as an excess linear-in-$T$ term in the crystal's specific heat, a test the paper itself suggests for future work.","The similarity to linear low-temperature broadening reported in other rare-earth crystals and in silicon-vacancy centres in diamond suggests a common defect-controlled mechanism; comparing crystals with different dopant concentrations and annealing histories could reveal whether the slope scales with defect density."],"forward_implications":["If the central claim holds, frequency references based on Eu:YSO spectral holes at dilution-refrigerator temperatures inherit a sub-kelvin linewidth that is temperature dependent, so the linear slope (roughly 0.5 to 1 kHz/K) must be included when predicting short-term stability.","The phase discriminant of $(0.64 \\pm 0.06)$ mrad/Hz means detection-noise rejection in a laser lock improves by more than a factor of two relative to operation at 3.5 K, without using a polarizing magnetic field.","Because all three measurement protocols give the same linear behaviour, the trend persists whether the hole is heated after burning or burned fresh at each temperature.","If the linear broadening comes from two-level systems, then crystal growth and annealing conditions become a handle on linewidth: reducing such defects should directly narrow holes at sub-kelvin temperatures."],"supporting_citations":[{"why":"Supplies the two-phonon Raman $T^7$ broadening law and one measured coefficient used as the baseline prediction.","marker":"[18]"},{"why":"Provides the larger $T^7$ coefficient and persistent-hole data at 1.2 K that the sub-kelvin slopes are compared against.","marker":"[23]"},{"why":"Reports linear dephasing by disorder modes in some Eu:YSO samples, the 'anomalous crystal' classification the authors argue does not apply to their sample.","marker":"[29]"},{"why":"Establishes sample-dependent optical dephasing attributed to two-level systems with approximately constant density of states.","marker":"[28]"},{"why":"Documents a linear linewidth temperature dependence below 1 K in silicon-vacancy centers in diamond, a comparable low-temperature observation.","marker":"[32]"},{"why":"Previous work on the same dilution-refrigerator setup showing sub-kelvin frequency shifts, the context for extending linewidth studies to this range.","marker":"[17]"}],"fun_headline_variants":["Sub-kelvin holes widen linearly, defying T^7","Linear linewidth rise in sub-kelvin optical emitters","Spectral holes broaden with heat at sub-kelvin","Eu:YSO holes: linear sub-kelvin broadening, not flat"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the measured spectral-hole width equals the true homogeneous linewidth at every temperature, with no residual power broadening, overburning, or instantaneous spectral diffusion whose contribution changes with temperature.","fun_headline_variants_meta":{"raw":{"variants":["Sub-kelvin holes widen linearly, defying T^7","Linear linewidth rise in sub-kelvin optical emitters","Spectral holes broaden with heat at sub-kelvin","Eu:YSO holes: linear sub-kelvin broadening, not flat"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000461,"raw_usage":{"total_tokens":2312,"prompt_tokens":954,"completion_tokens":1358,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":570,"completion_tokens_details":{"reasoning_tokens":1284}},"tokens_in":570,"tokens_out":1358,"duration_ms":13278,"temperature":1.0,"reasoning_tokens":1284,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T21:07:50.207547+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the homogeneous linewidth of the same Eu:YSO crystal between 100 mK and 1 K with a two-pulse photon-echo sequence that does not rely on hole burning: if the echo linewidth stays flat while the hole width climbs linearly with temperature, the linear slope is an artifact of the burning or probing protocol. A complementary check is a specific-heat measurement in the same range, since an excess linear-in-$T$ term would independently support the two-level-system interpretation.","supporting_citations":[{"cited_title":"Investissement d’Avenir","cited_arxiv_id":null,"evidence_quote":"Provides the larger $T^7$ coefficient and persistent-hole data at 1.2 K that the sub-kelvin slopes are compared against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports linear dephasing by disorder modes in some Eu:YSO samples, the 'anomalous crystal' classification the authors argue does not apply to their sample."},{"cited_title":"Zhang, N","cited_arxiv_id":null,"evidence_quote":"Establishes sample-dependent optical dephasing attributed to two-level systems with approximately constant density of states."}],"review_version":1}