{"id":"f1b95b76-e81b-4716-8618-1c9f54cd1bfb","arxiv_id":"2608.02118","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"First laboratory results from the Pollux UV test bench demonstrate working polarization generation, a first end-to-end polarimetric measurement, and a far-UV analyser with extinction ratio 10 at 120 nm.","lead":"This paper reports the first working results from a test bench built to calibrate the ultraviolet polarimeters of the Pollux instrument proposed for the Habitable Worlds Observatory. Initial tests show the bench can generate and measure polarized light in the UV, and a new mirror analyser reaches a ten-to-one extinction ratio in the far ultraviolet.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unmeasured residual polarisation from the integrating sphere (Sec. 2.1) could bias the theoretical reference states and the end-to-end validation; the paper provides no check at the 10^-3 level.","rationale":"The paper's strongest claim is modest: the bench operates and paves the way for characterisation of polarimetric precision, not that the 10^-3 precision has been achieved. The central support for this is the demonstrated ability to generate and measure known polarisation states. Every such demonstration in the MUV-NUV chain—the Block C validation against Mueller predictions (Sec. 2.3) and the end-to-end Stokes reconstruction (Sec. 2.4)—assumes that the light from Block A is fully depolarised. The integrating sphere is asserted to depolarise, but no measurement of the output Stokes vector is provided. For a bench whose entire purpose is to reach 10^-3 polarimetric precision, any residual polarisation at that level from the source or collimator would systematically bias the reference states and the validation results. This is the most load-bearing concern because it undermines the foundation of the bench's calibration, not just a single measurement. The other limitations (FUV analyser extinction ratio of 10, one unexplained depolarised order, lack of error bars) are either explicitly preliminary or localised and are flagged in the paper, whereas the unpolarised-source assumption is not flagged as a risk. The proposed null test is a direct, feasible check using existing hardware: by rotating the Block C Rochon and summing the two analyser outputs, one can isolate the source polarisation from the spectrometer's polarisation response. If the test shows modulation below 10^-3, the concern is resolved and the central claim stands; if above, the bench validation and the 'paving the way' statement would need qualification. This aligns with the reader's weakest_assumption and supports the reader's CONDITIONAL verdict, so no verdict change is needed.","tokens_in":5832,"tokens_out":9111,"duration_ms":76601,"concrete_test":"Perform a null test with the existing optics: set the Babinet-Soleil compensator to zero retardance (or remove it), place the Block C Rochon prism on a motorised rotation stage, and record the intensities of the S and P analyser orders at several wavelengths in the 138–290 nm range as the Rochon is stepped over 180°. For each wavelength, compute the summed intensity I_sum(θ)=I_S(θ)+I_P(θ) and fit I_sum(θ) = a0 + a2 cos(2θ+φ2) + a4 cos(4θ+φ4). If the peak-to-peak amplitude of the fitted modulation exceeds 10^-3 of a0 (the 10^-3 precision target), then the integrating-sphere output is not adequately depolarised and the Block A/B reference is biased; the bench validation and all absolute Stokes references in Figs. 4 and 6 must be revisited. If the modulation is below 10^-3 at all wavelengths, the unpolarised assumption is confirmed and the central claim stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sections 2.3 and 2.4 validate the MUV-NUV chain by comparing measured intensities and recovered Stokes vectors with Mueller-matrix predictions. Both comparisons assume the light entering Block C is fully depolarised, as stated in Sec. 2.1 ('an integrating sphere that depolarises the light'). If the sphere or collimator (Block B) imparts a residual polarisation—e.g., through a specular reflection from the sphere wall, non-Lambertian coating, or polarisation-dependent collimator efficiency in the VUV—then every 'theoretical' input Stokes vector is biased by that amount. The measured agreement in Fig. 4 and the Poincaré sphere reconstruction of Fig. 6 could then partially be a self-consistent error: the demodulation model uses Mueller matrices that lack the source term, and the unknown source polarisation would be absorbed into the inferred instrument response. For a bench whose stated purpose is to characterise polarimetric precision to 10^-3, an uncalibrated input polarisation at or above that level is a critical systematic. The paper presents no null test, no rotating-polarizer measurement of the sphere output, and no estimate of the residual degree of polarisation. This is not an internal inconsistency, but a missing calibration that is directly load-bearing for the central claim that the bench is operational and ready for precision characterisation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports first laboratory results from a dedicated vacuum-ultraviolet test bench for the Pollux spectropolarimeter proposed for HWO. In the MUV-NUV configuration (120-290 nm), the full optical chain is integrated: a polarized deuterium spectrum is obtained, the polarization-generation subsystem (Rochon prism + Babinet-Soleil compensator) is validated against Mueller-matrix intensity predictions, and a first end-to-end polarimetric measurement is made by recovering Stokes vectors at several spectral orders. In the FUV configuration (98-120 nm), a windowless deuterium source is characterized, a K-mirror modulator alignment scheme is implemented, and a MgF2/B4C mirror analyser is manufactured and tested, giving a polarization extinction ratio of 10 at 120 nm. The authors conclude that the bench is operational and paves the way for characterizing the polarimetric precision of the Pollux polarimeters, targeting 10^-3.","tokens_in":6116,"tokens_out":3796,"duration_ms":31052,"significance":"If the reported results hold, the paper represents a meaningful step toward raising the TRL of UV spectropolarimetry for HWO. The strengths are the construction and integration of a difficult VUV bench, the use of standard Mueller-matrix predictions as a validation tool (with no ad hoc fit to the claimed result), the detailed K-mirror alignment strategy adapted from MICADO, and the first measurement of a custom reflective FUV analyser. The paper is appropriately cautious in calling these 'first lab results' and in flagging the need for further synchrotron measurements. The main risk is the uncalibrated input polarization of the integrating-sphere source, which directly enters the theoretical Stokes vectors used for validation; the paper needs to close this gap before the bench can credibly claim readiness for 10^-3 polarimetric precision.","major_comments":[{"comment":"The validation of polarization creation and end-to-end demodulation assumes that the integrating sphere in Block A produces fully depolarized light ('an integrating sphere that depolarises the light'). No measurement, null test, or quantitative upper limit is provided for residual polarization from the sphere or from the two-mirror collimator in Block B. Since the bench target is 10^-3, any residual input DoP—whether from a non-Lambertian coating, a specular reflection inside the sphere, or polarization-dependent mirror reflectivity—biases every 'theoretical' Stokes vector and can be partially absorbed into the inferred instrument response. Please add a direct test of the sphere output (e.g., a rotating Rochron or polarizer as a null test) and report a bound on the input degree of polarization, or re-analyse the validation with an unknown source term.","section":"Sec. 2.1, Figs. 4 and 6"},{"comment":"One diffraction order shows 'significant depolarisation', with the arrow pointing from the surface of the Poincaré sphere toward the centre, attributed tentatively to 'insufficient separation between the S and P beams'. This outlier is not quantified, and its presence weakens the claim that the complete measurement chain is validated. Please provide a per-order table of measured versus theoretical Stokes parameters, explain the outlier (e.g., order overlap, incomplete prism separation, extraction artifact), or justify its exclusion. The impact of this order on the inferred polarimetric precision must be stated.","section":"Sec. 2.4, Fig. 6"},{"comment":"The validation plots show no error bars, no uncertainty estimates, and no goodness-of-fit metric. Statements such as 'good agreement' and 'validate our polarisation creation method' are not quantitatively supported. For a bench whose purpose is to measure 10^-3 polarimetric precision, the authors should report per-order residuals, photon-noise-limited uncertainties, and a metric such as RMS Stokes error or reduced chi-square. This is necessary to substantiate the readiness claim.","section":"Secs. 2.3-2.4, Figs. 4-6"}],"minor_comments":[{"comment":"Use consistent SI formatting: '1µm', '200µm', and '10−6 mbar' should be '1 μm', '200 μm', and '10^-6 mbar'. Also '1.6×10−5' would be clearer as '1.6x10^-5'.","section":"Throughout"},{"comment":"References [2] and [3] are incomplete ('et al., G.' and 'et al., M. L. G.'); reference [4] has no publication venue or year details. These should be completed for reproducibility.","section":"References"},{"comment":"The analyser extinction ratio is reported at a single wavelength, 120 nm, which the abstract correctly qualifies. The planned synchrotron campaign should be described in terms of wavelength coverage and the precision with which the extinction ratio will be determined across the FUV range.","section":"Sec. 3.4, Fig. 12"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest about the preliminary status of the bench, and the central difficulty is the unmeasured residual polarization of the integrating-sphere source. If the authors can add a null test or otherwise bound the input DoP at the 10^-3 level, the manuscript would be suitable for publication. The unresolved depolarised order in Sec. 2.4 should also be addressed before the end-to-end validation claim is accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a status report from an instrument-development program, and it delivers exactly what a status report should: first light, a validated polarization-generation chain, one end-to-end Stokes reconstruction, and a first FUV analyser measurement. It does not claim more than it shows, which I appreciate. The bench is real and the measurements are consistent with standard Mueller-matrix predictions.\n\nWhat's new: the integrated MUV-NUV chain working end-to-end, including the first polarised spectrum and a six-angle demodulation that recovers input Stokes vectors. That's a real milestone for the project. The FUV source characterisation and the K-mirror alignment adaptation from MICADO are useful engineering details. The analyser extinction ratio of 10 at 120 nm is preliminary but a genuine lab measurement.\n\nSoft spots: the quantitative case is thinner than the text implies. There are no error bars on the extracted Stokes vectors, and the one order with significant depolarisation is waved off with \"may indicate insufficient separation\" - that needs a real explanation before the bench is used for precision work. Bigger issue: the whole validation assumes the integrating sphere output is completely depolarised (Sec 2.1), and there is no null test or residual DoP measurement. If the sphere or collimator feeds a few x10^-3 of polarisation into Block C, the 'theoretical' Stokes vectors are biased and the demodulation comparison in Fig 6 is partly self-consistent. That is not fatal for a first-light paper, but it is load-bearing for the stated goal of characterising polarimetric precision to 10^-3. The reader's stress-test note lands.\n\nAlso minor: the FUV analyser result is one wavelength, one reflectometer, and the paper itself says a synchrotron campaign is needed. Fine as a first result, but don't let it be cited as a validated polarimeter.\n\nWho it's for: people working on UV instrumentation for HWO and anyone building VUV polarimetric test benches. It deserves a regular referee; an editor shouldn't desk-reject it. The right outcome is a conditional accept with requests for error bars, a sphere-output polarisation check, and at least one more order's anomaly.","headline":"A solid, honest first-light status report for a UV spectropolarimeter test bench; the bench works, but the 10^-3 precision claim is not yet supported.","tokens_in":6623,"tokens_out":1595,"would_cite":true,"duration_ms":13374,"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":"This paper reports that the Pollux vacuum-ultraviolet spectropolarimeter test bench is now operational in both its MUV-NUV and FUV configurations, with first end-to-end polarimetric measurements and a validated FUV mirror analyser, opening","keywords":["Spectropolarimetry","Ultraviolet","Far-ultraviolet","Pollux","Habitable Worlds Observatory","UV-mirror","Mueller matrix","polarimetric precision"],"falsifier":"Directly measure the polarisation state of the integrating sphere's output—for instance by rotating the Rochon prism without the compensator and comparing the two output intensities—and check whether any residual polarisation exceeds the 10^-3 target; non-zero residuals would invalidate the unpolarised reference and require re-validation of the bench.","tokens_in":5736,"feed_emoji":"🔭","tokens_out":6925,"duration_ms":52706,"temperature":0.7,"pith_summary":"The paper is trying to establish that a dedicated vacuum-ultraviolet test bench can validate the two polarimeter designs planned for the Pollux spectropolarimeter on the Habitable Worlds Observatory. First lab results show the MUV-NUV chain working end-to-end, from polarisation generation to Stokes reconstruction, and the FUV components (windowless deuterium source, K-mirror alignment, mirror analyser) individually working. If correct, the bench is ready for the next milestone: measuring polarimetric precision down to 10^-3, a requirement for the instrument's science case. This matters because laboratory proof of this kind is what moves UV spectropolarimetry from concept toward a real space mission.","feed_headline":"Pollux UV polarimeter bench works end-to-end","feed_subtitle":"MUV-NUV chain and FUV mirror analyser pass first checks; next goal is 10^-3 polarimetric precision.","key_machinery":"The central object is the two-configuration vacuum-ultraviolet test bench itself. In the MUV-NUV path, the key chain is: an integrating sphere that is assumed to fully depolarise the deuterium lamp beam, a Rochon prism plus a Babinet-Soleil compensator to synthesise arbitrary polarisation states, the CASSTOR modulator (two pairs of MgF2 retardation plates) followed by a Rochon analyser, and a cross-dispersed echelle spectrometer. In the FUV path, the central components are a windowless deuterium discharge source, a rotating K-mirror (a three-mirror assembly whose rotation changes polarisation without birefringent materials), and a multilayer analyser made of MgF2 over B4C on BK7. The Mueller","core_discovery":"The authors report that their vacuum-ultraviolet test bench has passed its first integrated laboratory tests. In the MUV-NUV configuration, the full optical chain was assembled: a first polarised deuterium-lamp spectrum was recorded, the polarisation-generation subsystem matched Mueller-matrix predictions across spectral orders, and a six-angle modulation/demodulation cycle recovered input Stokes vectors from created polarisation states. In the FUV configuration, a windowless deuterium plasma source was characterised and selected over nitrogen, a K-mirror alignment strategy was implemented, and a mirror-based analyser was manufactured and tested, showing a polarisation extinction ratio of 10","pith_inferences":["Inference: if the bench reaches its 10^-3 precision target, similar all-reflective FUV polarimeters could be validated in a standard vacuum laboratory rather than only at synchrotron facilities, lowering the cost of future instrument development.","Inference: the two-source K-mirror alignment method, which decouples four alignment degrees of freedom, is reusable for any reflective polarimeter that must stay aligned while sealed inside a vacuum chamber.","Inference: the integrating-sphere depolarisation assumption could be tested independently by placing a known analyser directly after the source block; if residual polarisation is found, an extra depolarising element or a calibration measurement would correct the reference.","Inference: the order-by-order Stokes reconstruction errors visible on the Poincaré sphere suggest where the future precision campaign should look first, such as checking S/P beam separation per order and its wavelength dependence."],"forward_implications":["The MUV-NUV bench can now generate, modulate, and demodulate polarisation states end-to-end, so the next step is a dedicated campaign to measure polarimetric precision.","The FUV analyser's measured 10:1 extinction at 120 nm is sufficient to attempt first FUV polarimetric measurements once the K-mirror and analyser are integrated inside the vacuum chamber.","Because deuterium feeds both the MUV-NUV lamp and the windowless FUV source, the two bench configurations can be cross-calibrated against each other.","The roughly balanced grating efficiency for the two orthogonal polarisations means the spectrometer does not introduce a large polarisation bias, simplifying data reduction.","Demonstrating bench operation raises the laboratory readiness of UV spectropolarimetry and prepares the ground for a future implementation on the Habitable Worlds Observatory."],"fun_headline_variants":["First light for Pollux UV polarimeter bench","Pollux polarimeter bench passes first lab tests","UV polarimeter bench shows first working rings","Pollux bench: full chain works, FUV analyzer hits 10:1","Pollux test bench: MUV-NUV chain, FUV mirror pass"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the integrating sphere in the light-source block fully depolarises the deuterium beam; if the sphere or collimator leaves any residual polarisation, the 'unpolarised' reference and the theoretical Stokes vectors used to validate the whole chain are systematically wrong.","fun_headline_variants_meta":{"raw":{"variants":["First light for Pollux UV polarimeter bench","Pollux polarimeter bench passes first lab tests","UV polarimeter bench shows first working rings","Pollux bench: full chain works, FUV analyzer hits 10:1","Pollux test bench: MUV-NUV chain, FUV mirror pass"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000247,"raw_usage":{"total_tokens":1421,"prompt_tokens":824,"completion_tokens":597,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":568,"completion_tokens_details":{"reasoning_tokens":524}},"tokens_in":568,"tokens_out":597,"duration_ms":5169,"temperature":1.0,"reasoning_tokens":524,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T14:49:56.345481+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly measure the polarisation state of the integrating sphere's output—for instance by rotating the Rochon prism without the compensator and comparing the two output intensities—and check whether any residual polarisation exceeds the 10^-3 target; non-zero residuals would invalidate the unpolarised reference and require re-validation of the bench.","supporting_citations":[],"review_version":1}