{"id":"bc28c057-194e-4060-842e-77cea348b3d8","arxiv_id":"2502.01212","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"With optics held by adjustable clamps on a Zerodur bench, a 10 cm cavity showed length noise below 1 pm per root hertz down to 3 mHz, meeting LISA ground-test requirements after under an hour of alignment.","lead":"Physicists in Hamburg built a laser-interferometer test bench whose mirrors sit in adjustable clamps on a low-expansion glass plate instead of being permanently glued, and showed the setup keeps length stable at the picometer level down to very low frequencies. The point is to cut the time to build and reconfigure ground-test interferometers for LISA and similar space missions from months to hours, at the same precision.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central pm-stability claim rests on a single non-stationary, detrended beat record with no run statistics; 'verified below 1 pm/√Hz down to 3 mHz' overstates what the paper establishes.","rationale":"The paper is a candid experimental demonstration, and the central measurement chain is sound in its main direction: a beat between a TAPSI prototype cavity and a previously characterized ULE reference cavity, converted through Eq. (2), with the PDH trace shown to be below the LISA requirement over most of the band. The reference cavity stability is more than an order of magnitude below the claimed limit, so it is not the limiting factor. The most load-bearing weakness is not the unmeasured mount CTE, which the reader highlighted, but the evidential status of the single non-stationary, detrended ASD that supports the 'verified below 1 pm/√Hz down to 3 mHz' statement. The paper's own text in Section 5 and Fig. 8 discloses non-stationary noise and drift removal; those disclosures justify treating the result as a best-achieved upper bound rather than a verified, reproducible property. This is consistent with the reader's CONDITIONAL verdict, so I recommend no change to the verdict. A reanalysis of the raw time series with segment statistics and an explicit detrending specification would settle the concern.","tokens_in":20271,"tokens_out":10811,"duration_ms":576099,"concrete_test":"Publish the raw PDH beat-note time series used for Fig. 7 and reprocess it end-to-end: (1) state exactly which linear and quadratic trends are removed; (2) split the record into at least five independent segments and compute the ASD at 3 mHz for each; (3) report median and spread with and without detrending. If the median at 3 mHz is above 1 pm/√Hz or the segment-to-segment spread exceeds about 30%, the abstract should be softened to 'best-achieved upper bound under favorable conditions.'","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 5 (Figs. 7 and 8) reports the PDH cavity-length ASD from a beat-note time series that contains a linear temperature drift (Eq. 8) and a quadratic drift, both removed before spectral estimation. The residual after detrending is explicitly described as containing 'non-stationary noise' attributed to parasitic beams. The abstract asserts 'verified that the cavity length noise is below 1 pm/sqrt(Hz) for frequencies down to 3 mHz,' but the paper gives no run-to-run statistics, no uncertainty bars on the ASD, and no statement of how detrending affects the 3 mHz point. The presented trace is therefore a favorable-case upper bound on the cavity length noise, not a verified property of the TAPSI concept. This is not an accusation that the result is wrong: the total measured noise is still a valid upper bound, and readout and stray light likely make the cavity look worse than it is. The concern is that the headline wording overstates what a single non-stationary record can establish. The unmeasured ZeroDrift CTE and compensation-plate matching (Sec. 2) are a separate generalization risk; the more immediate risk is that the low-frequency endpoint is not reproducible across independent records.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents TAPSI, an opto-mechanical concept in which mirrors and other optics are placed on a Zerodur bench with adjustable, clamped Invar posts and thermally compensated mirror mounts, rather than being bonded or glued. A 10 cm prototype cavity built with this concept is characterized against a ULE reference cavity using two locking schemes, heterodyne stabilization (HS) and Pound-Drever-Hall (PDH) locking. The authors report a temperature stability of 10 µK/√Hz down to 10 mHz, an initial alignment time under one hour, and a PDH cavity-length noise that they state is below the LISA requirement for most frequencies and only slightly above at 0.8–3 mHz. The abstract goes further, claiming that the cavity length noise is verified to be below 1 pm/√Hz down to 3 mHz. The paper concludes that TAPSI is suitable for flexible optical ground support equipment for LISA and future missions.","tokens_in":20307,"tokens_out":6202,"duration_ms":51923,"significance":"If the central claim holds, TAPSI would be a practically valuable alternative to permanently bonded optical benches for ground-based testing of space interferometers: it promises reconfigurability and much faster assembly while maintaining picometer-level length stability. The paper has notable strengths: the length measurement is made against an external reference cavity, the thermal floor of the Zerodur bench is estimated forward from measured temperature noise and literature CTE values, two locking schemes are compared, and the authors are candid about known limitations such as stray light and the unmeasured mount CTE. However, the headline 'verified below 1 pm/√Hz down to 3 mHz' is not fully supported by the evidence presented, and the paper's own Section 5 appears to contradict the abstract at the 3 mHz point. The underlying concept is credible and the measurement is a useful upper bound, but the verification claim needs either additional data or more careful wording.","major_comments":[{"comment":"The abstract states that 'we verified that the cavity length noise is below 1 pm/sqrt(Hz) for frequencies down to 3 mHz,' but Section 5 says the PDH trace 'is below the LISA requirements for most frequencies and only slightly above at 0.8 to 3 mHz.' Since Eq. (1) gives u(3 mHz) ≈ 1.09 pm/sqrt(Hz), being above the LISA requirement at 3 mHz implies the noise is not below 1 pm/sqrt(Hz) at that frequency. Please reconcile these statements by specifying the exact frequency at which the trace crosses 1 pm/sqrt(Hz) and clarify whether the claim applies to 3 mHz inclusive.","section":"Abstract; Section 5; Eq. (1)"},{"comment":"The central verification rests on a single beat-note ASD. Figure 8 shows a linear temperature drift removed via Eq. (8) and a quadratic drift removed before spectral estimation, and the residual is described as containing non-stationary noise attributed to parasitic beams. The paper gives no run-to-run statistics, no uncertainty bars on the ASD, and no analysis of how the detrending affects the value at 3 mHz. The plotted trace is therefore a favorable processed upper bound from one non-stationary record, not a statistical verification of the stated stability. Please provide at least two independent records, error bars or segment-to-segment scatter, and a statement of the detrending procedure's effect at the lowest reported frequency.","section":"Section 5, Figs. 7 and 8, Eq. (8)"},{"comment":"The thermal-compensation mechanism is central to the design: the mirror mount and Invar compensation plate are intended to cancel each other's expansion, yet the CTE of the ZeroDrift mount is stated to be unknown and the authors note that a future study is needed to optimize the plate. The demonstrated picometer stability is therefore shown for one specific geometry and thermal environment, not as a property of the adjustable mounting concept in general. Either measure the mount's effective CTE or demonstrate repeatability after reassembly or in a different configuration before claiming that TAPSI provides adjustable picometer-stable interferometers as a general toolset.","section":"Section 2"}],"minor_comments":[{"comment":"Equation (8) is typographically ambiguous: 'dL/dT = d f/fLaser LCavity · 1/dT' should be written as (df/f_laser) * L_cavity / dT, with units specified for each quantity.","section":"Section 5, Eq. (8)"},{"comment":"The phrase '7.5 fm/sqrt(Hz) (20 Hz sqrt(Hz)) down to 3 mHz' appears malformed; please clarify whether the reference cavity stability is 7.5 fm/sqrt(Hz) at 20 Hz or some other specification.","section":"Section 4"},{"comment":"References [20] and [21] are the same paper (Korth et al., 'Passive, free-space heterodyne laser gyroscope'); one should be removed and citations renumbered.","section":"References"},{"comment":"The abbreviation list contains 'ASD' twice; please deduplicate.","section":"List of Abbreviations"},{"comment":"The caption mentions 'the free-running laser noise of a 10 cm cavity' but the text does not define how this curve was measured or modeled; please add a brief explanation.","section":"Figure 7 caption"},{"comment":"The text refers to 'Johnsen-Nyquist noise'; this should be 'Johnson-Nyquist noise.'","section":"Section 3"}],"recommendation":"major_revision","confidential_remarks":"The paper reports a useful and plausible demonstration of an adjustable, non-bonded optical bench for ground-based space-interferometer testing. The main gap is not the measurement principle but the strength of the verification claim: a single detrended, non-stationary record is used to support an abstract statement of verification below 1 pm/sqrt(Hz) down to 3 mHz, and that statement appears inconsistent with the body of the paper. I recommend asking the authors to either add the missing statistical support (independent records, error bars, detrending sensitivity) or soften the claim to an upper-bound demonstration. The unmeasured CTE of the mirror mounts is an important scope limitation but not a reason to reject if the paper is framed as a proof-of-principle for one configuration."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things worth knowing. The headline result is real but slightly oversold. The paper shows a 10 cm cavity built from freely positionable clamped mounts on a Zerodur bench reaching length noise below 1 pm/√Hz down to about 3 mHz, with alignment in under an hour. That is a genuinely useful result for LISA ground testing. But the abstract's \"verified\" overstates what the body actually shows: the PDH trace is below the LISA requirement for most frequencies and only slightly above at 0.8–3 mHz, and the low-frequency endpoint comes from a single detrended record with non-stationary noise.\n\nWhat's new: it extends Kulkarni et al. by freeing the mount positions, adding an Invar compensation plate, and building a purposefully isolated thermal facility. The demonstration is candid. They document the HS delay-beat limitation, RAM from fiber circulators, stray light, PD heating, and explicitly flag the non-stationarity. The measurement chain is sound: beat against a ULE reference cavity, conversion via Eq. (2), and the bench thermal floor is computed forward from literature CTE and measured temperature noise. The one fitted quantity, dL/dT, is used diagnostically and does not carry the result. Since readout and stray light make the cavity look worse than it is, the trace is a legitimate upper bound.\n\nSoft spots, in proportion. The central claim rests on one record: no run-to-run statistics, no error bars on the ASD, and the quadratic detrending is not quantified for the 3 mHz point. That is a moderate concern for the wording, not for the physics; the result is probably right. Second, the CTE of the ZeroDrift mounts is never measured, and the compensation plate is not thermally cycled. The cancellation between mount and plate is the mechanism that supposedly makes the assembly stable, so the generality of the concept is asserted rather than demonstrated. That is a real but minor gap for the specific geometry shown. Third, the reference cavity stability is taken from a 2005 thesis rather than re-measured; acceptable as a benchmark, but it should be stated as an assumption rather than an external ground truth.\n\nWho this is for: teams building optical ground support equipment for LISA, Taiji, TianQin, or GRACE-FO-class missions, and anyone who wants a faster way to prototype picometer-stable interferometers. It deserves a serious referee. The revision requests are straightforward: soften the abstract, add at least one repeat measurement or error bars, and ideally one thermal-cycling test of the mount assembly.\n\nSend it to review.","headline":"A genuinely useful engineering demonstration of reconfigurable picometer-stable interferometers, but the abstract's 'verified' overstates what a single non-stationary, detrended beat record can establish.","tokens_in":21089,"tokens_out":2094,"would_cite":true,"duration_ms":19321,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A clamp-mounted optical bench keeps cavity length noise below 1 pm/√Hz down to 3 mHz, reaching LISA ground-test stability for most frequencies.","keywords":["laser interferometry","gravitational wave detectors","LISA","optical ground support equipment","ultra-stable optical bench","Pound-Drever-Hall locking","heterodyne frequency stabilization","picometer length stability"],"falsifier":"Directly measure the coefficient of thermal expansion of the ZeroDrift mirror-mount and Invar compensation-plate assembly, then compute the resulting 10 cm cavity length drift; if the drift projected onto the noise spectrum rises above the LISA requirement curve in the 0.1-3 mHz band, the assumed thermal cancellation fails.","tokens_in":19849,"feed_emoji":"🔭","tokens_out":6758,"duration_ms":55433,"temperature":0.7,"pith_summary":"The paper argues that a laser interferometer used for ground-testing space-based gravitational wave detectors does not have to be a permanently bonded assembly. The authors built a toolset, TAPSI, in which optics are clamped and screwed onto a Zerodur plate, freely repositionable like components on an optical table. Locking a laser to a 10 cm cavity built this way, they measured cavity length noise below $1\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ down to $3\\,\\mathrm{mHz}$ using Pound-Drever-Hall locking, below the LISA requirement for most frequencies. That means a given interferometer configuration can be set up and aligned in under an hour instead of weeks or months, which would make flexible optical ground support equipment practical.","feed_headline":"Adjustable optics hold picometer stability down to 3 mHz","feed_subtitle":"A clamped, freely positioned mirror setup on a Zerodur bench meets LISA requirements in hours, not weeks.","key_machinery":"The central object is the toolset for adjustable picometer-stable interferometers (TAPSI): a Zerodur optical bench on which optics sit on posts held by clamps and screws, with Newport ZeroDrift mirror mounts chosen for low tilt-to-length coupling and an Invar compensation plate placed between mount and post so that the thermal expansions of mount and plate cancel at the mirror center. The argument is carried by measuring the length change of a 10 cm prototype cavity built with these mounts, comparing its beat frequency against a 21 cm ULE reference cavity with known $7.5\\,\\mathrm{fm}/\\sqrt{\\mathrm{Hz}}$ stability, using two locking schemes (heterodyne stabilization and Pound-Drever-Hall). The PDH result is the carrier of the central claim: it stays below the LISA requirement curve except for a slight excess at 0.8-3 mHz.","core_discovery":"The central demonstrated claim is that adjustable, clamp-mounted optics on a low-expansion Zerodur bench can be picometer-stable at the low frequencies relevant to LISA. Using a prototype 10 cm cavity with finesse 627 assembled entirely with the TAPSI concept, a 1064 nm laser locked by the Pound-Drever-Hall technique showed relative cavity length noise below $1\\,\\mathrm{pm}/\\sqrt{\\mathrm{Hz}}$ down to $3\\,\\mathrm{mHz}$, below the LISA requirement curve except for a small excess between 0.8 and 3 mHz. The same cavity probed with heterodyne laser frequency stabilization was limited by an analogue demodulation delay effect, not by the mounting concept. The authors therefore conclude that the concept is suitable for flexible, reusable optical ground support equipment for LISA and future space interferometry missions.","pith_inferences":["If the mount-plate thermal cancellation holds more generally, the concept could be portable to other low-expansion benches and to testbeds where bonding is impractical; the paper only demonstrates one cavity geometry, so this portability is an inference.","The unmeasured coefficient of thermal expansion of the ZeroDrift mounts means the compensation plate is currently matched by assumption; a direct CTE characterization would let the design be tuned, and thermal cycling tests would reveal whether the stability is robust across temperature ranges.","The sub-hour assembly time suggests a workflow where candidate interferometer layouts for LISA ground support are prototyped and compared on one bench before a final bonded design is committed; the paper does not explicitly propose this workflow.","Because the heterodyne approach was limited by an analogue demodulation delay rather than by the optics, moving to digital demodulation could let the simpler heterodyne scheme itself reach LISA stability on the same bench; the paper plans this but has not demonstrated it."],"forward_implications":["Interferometers for ground testing of LISA can be assembled in hours rather than weeks or months, because components are clamped and repositionable rather than bonded.","The same bench and components can be reused across different experiments, reducing the cost of optical ground support equipment.","The concept extends to other future low-frequency space interferometry missions such as Taiji and DECIGO, which need picometer-stable ground testbeds.","The remaining gap at 0.8-3 mHz in the PDH lock is attributed to stray light and readout effects rather than the mounting concept, so higher cavity finesse and digital demodulation should close it."],"supporting_citations":[{"why":"Provides the prior adjustable-mirror scheme on a low-expansion baseplate that TAPSI extends to freely positionable arbitrary configurations.","marker":"[22]"},{"why":"Supplies the thermally compensated ZeroDrift mirror mounts whose low tilt-to-length coupling is central to the mounting concept.","marker":"[27]"},{"why":"Introduces the Pound-Drever-Hall locking method used to measure the prototype cavity length stability.","marker":"[5]"},{"why":"Describes the heterodyne laser frequency stabilization scheme used as the second locking method and identifies the digital demodulation fix for the delay-beat limitation.","marker":"[12]"},{"why":"Provides the reference cavity with known $7.5\\,\\mathrm{fm}/\\sqrt{\\mathrm{Hz}}$ stability used as the baseline for beat-frequency measurements.","marker":"[37]"},{"why":"Supplies the LPSD spectrum estimation method used to compute the amplitude spectral densities of the length and temperature noise.","marker":"[38]"},{"why":"Represents the bonded optical bench approach for LISA that the adjustable concept is intended to complement or replace.","marker":"[6]"},{"why":"Demonstrates the digital-domain frequency-dependent phase compensation that overcomes the delay-beat limiting the heterodyne stabilization.","marker":"[7]"}],"fun_headline_variants":["Picometer-stable adjustable optics built in hours","Adjustable mounts hit LISA-grade stability at 3 mHz","Clamp-on optics achieve sub-picometer noise to 3 mHz","Rapid assembly, picometer stability for LISA testbeds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole approach rests on the assumption that the thermal expansion of the mirror mounts and the Invar plate underneath them cancel well enough to keep the mirror position stable, but that cancellation is never measured directly in the paper.","fun_headline_variants_meta":{"raw":{"variants":["Picometer-stable adjustable optics built in hours","Adjustable mounts hit LISA-grade stability at 3 mHz","Clamp-on optics achieve sub-picometer noise to 3 mHz","Rapid assembly, picometer stability for LISA testbeds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000194,"raw_usage":{"total_tokens":1390,"prompt_tokens":1019,"completion_tokens":371,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":635,"completion_tokens_details":{"reasoning_tokens":298}},"tokens_in":635,"tokens_out":371,"duration_ms":4236,"temperature":1.0,"reasoning_tokens":298,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T16:02:22.175125+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly measure the coefficient of thermal expansion of the ZeroDrift mirror-mount and Invar compensation-plate assembly, then compute the resulting 10 cm cavity length drift; if the drift projected onto the noise spectrum rises above the LISA requirement curve in the 0.1-3 mHz band, the assumed thermal cancellation fails.","supporting_citations":[{"cited_title":"Ultrastable optical components using adjustable commercial mirror mounts anchored in a ule spacer","cited_arxiv_id":null,"evidence_quote":"Provides the prior adjustable-mirror scheme on a low-expansion baseplate that TAPSI extends to freely positionable arbitrary configurations."},{"cited_title":"Suprema ® zerodrift™ thermally compensated mirror mounts - su100tw-f2k","cited_arxiv_id":null,"evidence_quote":"Supplies the thermally compensated ZeroDrift mirror mounts whose low tilt-to-length coupling is central to the mounting concept."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the Pound-Drever-Hall locking method used to measure the prototype cavity length stability."},{"cited_title":"Tanner, and Guido Mueller","cited_arxiv_id":null,"evidence_quote":"Describes the heterodyne laser frequency stabilization scheme used as the second locking method and identifies the digital demodulation fix for the delay-beat limitation."},{"cited_title":"Laser development and stabilization for the spaceborne interferometric gravitational wave detector lisa","cited_arxiv_id":null,"evidence_quote":"Provides the reference cavity with known $7.5\\,\\mathrm{fm}/\\sqrt{\\mathrm{Hz}}$ stability used as the baseline for beat-frequency measurements."},{"cited_title":"Improved spectrum estimation from digitized time series on a logarithmic frequency axis","cited_arxiv_id":null,"evidence_quote":"Supplies the LPSD spectrum estimation method used to compute the amplitude spectral densities of the length and temperature noise."},{"cited_title":"Bogenstahl, C","cited_arxiv_id":null,"evidence_quote":"Represents the bonded optical bench approach for LISA that the adjustable concept is intended to complement or replace."},{"cited_title":"Integrating high-precision and fringe-scale displacement sensing using heterodyne cavity- tracking","cited_arxiv_id":null,"evidence_quote":"Demonstrates the digital-domain frequency-dependent phase compensation that overcomes the delay-beat limiting the heterodyne stabilization."}],"review_version":1}