Model of the instrumental polarization induced by a rotating HWP with anti-reflective sub-wavelength structures for CMB polarimetry
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics. The day in astrophysics: observations, models and what the data forces.
Vera: We're looking at the paper: "Model of the instrumental polarization induced by a rotating HWP with anti-reflective sub-wavelength structures for CMB polarimetry".
Jocelyn: Here's the gist of the paper. Future CMB polarization experiments require precise control of instrumental systematics in the search for primordial B-mode polarization <ref:2610.38165</ref:1> The results suggest that improving the total transmission, including the effect of the fabricated SWS profile, may be more important for reducing HWPinduced instrumental polarization than tuning the SWS asymmetry itself <ref:2610.38165</ref:24>.
Vera: System Modeling and Methodology The study employs rigorous coupled-wave analysis (RCWA) to characterize the response of a multi-layer sapphire HWP with anti-reflective sub-wavelength structures (SWS) <ref:2610.38165</ref:7>.
Jocelyn: The authors define coordinate systems and use Mueller calculus to model the HWP and detector <ref:2610.38165</ref:22>.
Vera: They construct a HWP model composed of five birefringent flat plates with asymmetric pyramidal SWS on the external faces of plates 1 and 5 <ref:2610.38165</ref:9>.
Paper summary: Vera: Okay, so we're looking at this paper, "Model of the instrumental polarization induced by a rotating HWP with anti-reflective sub-wavelength structures for CMB polarimetry". This whole thing is about getting those precise measurements of the primordial B-mode polarization we need for future experiments.
Jocelyn: Right. The paper focuses on how a rotating half-wave plate, which we use to modulate the polarization, can introduce its own instrumental errors, and it tests how adding sub-wavelength structures helps control that.
Subrahmanyan: So the core idea here is tackling these instrumental systematics because they really muddy the water when we try to find that tiny primordial signal from inflation two <ref:2610.38165#pg1>. It’s about controlling what we measure so we can actually see what happened in the very early universe.
Vera: Exactly. The authors are using rigorous coupled-wave analysis to characterize a multi-layer sapphire half-wave plate with these anti-reflective sub-wavelength structures, or SWS, to see how it reacts when rotated seven <ref:2610.38165#pg2>. They test three different asymmetric SWS models and find that the type of asymmetry doesn't really make a big difference in transmittance or the polarization efficiency twenty-four <ref:2610.38165#pg2>.
Jocelyn: That’s interesting because we often think tuning the structure itself is key, but they found that for these three models, it doesn't substantially change how much instrumental polarization you get seven <ref:2610.38165#pg2>. What they did show is that if you compare their idealized model to one with a realistic SWS geometry—one where the asymmetry enhances the retardance—higher transmittance seems to lead to lower instrumental polarization seventeen <ref:2610.38165#pg2>.
Subrahmanyan: That suggests that improving the total transmission, including how the actual fabricated structure behaves, might be more important for reducing that HWP-induced polarization than just tweaking the shape of those sub-wavelength structures themselves twenty-four <ref:2610.38165#pg2>. It shifts the focus from just geometry to overall optical performance.
Vera: It does make sense. They are looking at a fourth-harmonic instrumental polarization, and they note that this is typically around ten to the power of negative four in CMB bands, which is something we expect to be manageable if we can control it four <ref:2610.38165#pg2>. This level should be treatable with some mitigation methods already proposed by Patanchon et al. twenty-two <ref:2610.38165#pg1>.
Jocelyn: And they also looked at how things change depending on the incident angle and frequency, noting that the phase doesn't really depend on the angle up to a certain precision in their simulations thirteen <ref:2610.38165#pg2>. They did look at color effects too, checking if dust and synchrotron emissions mess things up, but they found those changes in band-averaged efficiency are only at the percent level or below for most of our frequency bands twenty <ref:2610.38165#pg1>.
Subrahmanyan: The theoretical side is interested because this instrumental polarization is what directly contributes to the reconstructed CMB polarization, which is what we’re aiming for four <ref:2610.38165#pg2>. If we can keep this noise down, it opens up access to the physics of inflation and B-mode signals that haven't been detected yet
three–thirteen: <ref:2610.38165#pg2>.
Vera: So they’ve characterized this specific component of the error by decomposing the Mueller coefficients into harmonics of the half-wave plate rotation angle, showing that instrumental polarization lives in those fourth-harmonic coefficients four <ref:2610.38165#pg2>. This is a concrete way to isolate exactly what we need to worry about.
Jocelyn: And when they looked at the actual physical setup, they found that this instrumental polarization corresponds to an additional polarization angle rotation of about thirty arcseconds at its maximum value twenty-seven <ref:2610.38165#pg2>. That’s a specific angular shift we need to account for in our data model if we want to get the signal right.
Subrahmanyan: That thirty arcsecond rotation is a specific effect that needs modeling, but as long as it stays around ten to the power of negative four, it’s something we can handle with existing techniques like those from Patanchon et al. twenty-two <ref:2610.38165#pg1>. The crucial part for us is making sure this signal is characterized and included correctly in our time-ordered data model twenty-four <ref:2610.38165#pg2>.
Vera: So to wrap up what the paper did, they showed that while tuning the asymmetry of the SWS doesn't give you a huge win, improving the total transmission seems like a better path to reducing HWP-induced instrumental polarization seventeen <ref:2610.38165#pg2>. They are setting up this framework for how we need to model these errors when we look for those primordial B-modes.
Jocelyn: It’s about building the tools that let us separate the true cosmic signal from the noise introduced by our hardware, and this paper gives us a concrete way to understand that noise source. It shows exactly where we should be focusing our efforts in designing better polarization instruments for CMB experiments.
Subrahmanyan: Ultimately, understanding this HWP-induced instrumental polarization is a necessary step before we can truly hope to detect the faint B-mode signal that tells us about inflation two <ref:2610.38165#pg1>. It’s a practical engineering constraint that has cosmological implications.
Conclusion: Vera: So we're wrapping up this look at how a rotating half-wave plate interacts with these sub-wavelength structures for CMB polarimetry.
Jocelyn: It’s about this paper, "Model of the instrumental polarization induced by a rotating HWP with anti-reflective sub-wavelength structures for CMB polarimetry." The authors are showing us exactly what those errors look like when you're trying to measure the faint B-mode polarization from the early universe.
Subrahmanyan: They use rigorous coupled wave analysis, or RCWA, to figure out how this sapphire HWP model responds across different configurations of those anti-reflective sub-wavelength structures.
Vera: And they find that the instrumental polarization—the noise created by the hardware itself—is mainly in the fourth harmonic component, which is usually around ten to the power of negative four in those CMB bands.
Jocelyn: They focus on this fourth harmonic because it's what directly feeds into our reconstructed CMB polarization signal.
Subrahmanyan: The key finding here is that tuning the asymmetry of those sub-wavelength structures doesn't really change the overall level of instrumental polarization significantly across the three models they tested.
Vera: But there’s a nuance they found, which is interesting for us on the ground. When you compare their idealized model to one with a more realistic geometry—one where the asymmetry actually enhances how much light is transmitted—higher transmittance seems to lead to lower instrumental polarization.
Jocelyn: So, it suggests that optimizing the total transmission of the system might be a better way to keep that noise down than just fine-tuning the shape of those sub-wavelength structures themselves.
Subrahmanyan: That makes sense from a structural standpoint because we're dealing with physical fabrication constraints, and if we can get better overall transmission, that’s a practical win for reducing these systematic errors.
Vera: And they also gave us some specific numbers, like an additional polarization angle rotation of about thirty arcseconds at its maximum when the system is rotating.
Jocelyn: That thirty arcsecond shift is a concrete piece of data; it tells us exactly what kind of additional polarization we need to model in our time-ordered data when we look at the detector signal.
Subrahmanyan: So, they've set up a clear path forward by characterizing this noise source so that future experiments can incorporate it into their data models properly.
Vera: It’s about moving from just guessing the error to actually knowing what the hardware is doing and how we can subtract it out of the signal.
Jocelyn: And that characterization, especially at this level of precision, is what will let us finally start seeing those primordial B-modes in our next generation of CMB experiments.
Okayama University · National Astronomical Observatory of Japan · Université Paris-Cité, CNRS, Astroparticule et Cosmologie, Paris F-75013, France · Suwa University of Science
astro-ph.IM, astro-ph.CO
Submitted: 2026-10-28
Updated: 2026-11-11
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 83/100
The gist: Future CMB polarization experiments require precise control of instrumental systematics in the search for primordial B-mode polarization <ref:2610.38165</ref:1> The results suggest that improving the…
Key concepts
- Rigorous Coupled-Wave Analysis (RCWA)
- This is a mathematical technique used to precisely calculate how electromagnetic waves interact with complex structures like multi-layer sapphire HWP with SWS. It allows researchers to model the exact light transmission and polarization response under various conditions, providing a detailed physical picture of the system's behavior.
- Instrumental Polarization (IP)
- This is unwanted polarization introduced by the instrument itself, specifically caused by imperfections in components like a rotating HWP. The paper focuses on how this unwanted signal manifests as specific harmonics (like the fourth harmonic) and how to reduce it for accurate CMB measurements.
- Anti-Reflective Sub-wavelength Structures (SWS)
- These are tiny structures placed on the surface of the sapphire HWP designed to minimize reflection. The study investigates different asymmetric SWS models, finding that while asymmetry doesn't drastically change performance metrics, optimizing the overall transmission through these structures is key to lowering instrumental polarization.
Terminology
Summary
Future CMB polarization experiments require precise control of instrumental systematics in the search for primordial B-mode polarization <ref:2610. p. 038165</ref:1> The results suggest that improving the total transmission, including the effect of the fabricated SWS profile, may be more important for reducing HWPinduced instrumental polarization than tuning the SWS asymmetry itself <ref:2610. p. 038165</ref:24>.
System Modeling and Methodology
The study employs rigorous coupled-wave analysis (RCWA) to characterize the response of a multi-layer sapphire HWP with anti-reflective sub-wavelength structures (SWS) <ref:2610.38165</ref:7>. The authors define coordinate systems and use Mueller calculus to model the HWP and detector <ref:2610. p. 438165</ref:22>. They construct a HWP model composed of five birefringent flat plates with asymmetric pyramidal SWS on the external faces of plates 1 and 5 <ref:2610. p. 738165</ref:9>. Three types of asymmetric SWS models are investigated to see how different types of SWS asymmetry affect the level of instrumental polarization <ref:2610. p. 738165</ref:7>.
Harmonic Analysis and Instrumental Polarization
The instrumental polarization is dominated by the second and fourth harmonics, with amplitudes typically ranging from 10−4 to 10−2 and from 10−6 to 10−3 <ref:2610.38165</ref:4>. The focus is primarily on the fourth-harmonic component because it directly contributes to the reconstructed CMB polarization <ref:2610. p. 038165</ref:4>. The amplitude of this instrumental polarization is typically of order 10−4 in the CMB bands <ref:2610. p. 038165</ref:4>.
Dependence on Incident Angle and Frequency
The dependence of the Mueller matrix components on the incident angle is described by an empirical form M(2m)XY (Θ) = A + BΘ squared + CΘ 4 + DΘ 6 <ref:2610.38165</ref:13>. The phase is independent of the incident angle Θ up to the precision of simulations <ref:2610. p. 138165</ref:4>. The band-averaged amplitudes and phases are consistent with the relations expected from the theoretical symmetry of the fourth-harmonic terms <ref:2610. p. 038165</ref:13>.
Comparison of SWS Models and Performance Metrics
The comparison among the SWS models shows that the size of the instrumental polarization is not mainly determined by the asymmetry model itself <ref:2610. p. 0 p. 0 p. 038165</ref:24>. By comparing with the SWS model with realistic model, higher transmittance represented by M(0)II can lead to lower instrumental polarization <ref:2610. p. 0 p. 0 p. 0 p. 038165</ref:24>. Model-A remains attractive because it can reduce the required sapphire thickness while keeping comparable optical performance <ref:2610. p. 338165</ref:24>. The transmittance and IP appear to be anti-correlated, in higher transmittance is associated to lower IP <ref:2610. p. 038165</ref:17>.
Systematic Error Assessment
The study also evaluates the effect of color effects and foregrounds on the band-averaged response using representative spectral indices for dust and synchrotron emissions <ref:2610.38165</ref:20>. The difference of the amplitude ε(4)IQ − ε(4)IU and ε(4)QI −ε(4)UI is less than 2. p. 20×10−7, while ε(4)VQ −ε(4)UV and ε(4)QV −ε(4)UV is less than 2. p. 20×10−6 <ref:2610. p. 038165</ref:13>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>.
Conclusion
By decomposing the Mueller coefficients into harmonics of the HWP rotation angle, the instrumental polarization is in the fourth-harmonic Mueller coefficients <ref:2610.38165</ref:4>. This level is expected to be treatable with the mitigation method proposed by Patanchon et al. p. 3 p. 3 p. 3 p. 3 p. 3 [22], provided that the HWP-induced signal is characterized and included in the time-ordered data model <ref:2610. p. 2 p. 2 p. 2 p. 2 p. 238165</ref:24>. The corresponding additional polarization-angle rotation is approximately 30 arcsec at maximum <ref:2610. p. 6 p. 6 p. 6 p. 6 p. 638165</ref:27>. The foreground spectral indices within representative ranges changes the band-averaged efficiency only at the percent level or below for most frequency bands <ref:2610. p. 7 p. 7 p. 7 p. 7 p. 738165</ref:20>. The effect of the difference with the band coverage is also negligible if the instrumental polarization does not need to be modeled at better than the percent level <ref:2610.38165</ref:24>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>. The comparison among the SWS models shows that the size of the instrumental polarization is not mainly determined by the asymmetry model itself <ref:2610.38165</ref:24>. By comparing with the SWS model with realistic model, higher transmittance represented by M(0)II can lead to lower instrumental polarization <ref:2610.38165</ref:17>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>. The comparison among the SWS models shows that the size of the instrumental polarization is not mainly determined by the asymmetry model itself <ref:2610.38165</ref:24>. By comparing with the SWS model with realistic model, higher transmittance represented by M(0)II can lead to lower instrumental polarization <ref:2610.38165</ref:17>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>. By decomposing the Mueller coefficients into harmonics of the HWP rotation angle, we evaluated the HWP-synchronous components relevant to the detector signal <ref:2610. p. 4 p. 4 p. 438165</ref:4>. The instrumental polarization is in the fourth-harmonic Mueller coefficients <ref:2610. p. 0 p. 0 p. 0 p. 038165</ref:4>. Its amplitude is typically of order 10−4 in the CMB bands <ref:2610. p. 0 p. 0 p. 0 p. 038165</ref:4>. This level is expected to be treatable with the mitigation method proposed by Patanchon et al. [22], provided that the HWP-induced signal is characterized and included in the time-ordered data model <ref:2610.38165</ref:24>. The corresponding additional polarization-angle rotation is approximately 30 arcsec at maximum <ref:2610.38165</ref:27>. The foreground spectral indices within representative ranges changes the band-averaged efficiency only at the percent level or below for most frequency bands <ref:2610.38165</ref:20>. The effect of the difference with the band coverage is also negligible if the instrumental polarization does not need to be modeled at better than the percent level <ref:2610.38165</ref:24>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>. The comparison among the SWS models shows that the size of the instrumental polarization is not mainly determined by the asymmetry model itself <ref:2610.38165</ref:24>. By comparing with the SWS model with realistic model, higher transmittance represented by M(0)II can lead to lower instrumental polarization <ref:2610.38165</ref:17>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>. By decomposing the Mueller coefficients into harmonics of the HWP rotation angle, we evaluated the HWP-synchronous components relevant to the detector signal <ref:2610.38165</ref:4>. The instrumental polarization is in the fourth-harmonic Mueller coefficients <ref:2610.38165</ref:4>. Its amplitude is typically of order 10−4 in the CMB bands <ref:2610.38165</ref:4>. This level is expected to be treatable with the mitigation method proposed by Patanchon et al. [22], provided that the HWP-induced signal is characterized and included in the time-ordered data model <ref:2610.38165</ref:24>. The corresponding additional polarization-angle rotation is approximately 30 arcsec at maximum <ref:2610.38165</ref:27>. The foreground spectral indices within representative ranges changes the band-averaged efficiency only at the percent level or below for most frequency bands <ref:2610.38165</ref:20>. The effect of the difference with the band coverage is also negligible if the instrumental polarization does not need to be modeled at better than the percent level <ref:2610.38165</ref:24>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>.
A Modeling SWS
The authors model asymmetric AR-SWS using the ellipse contour along the z-axis <ref:2610.38165</ref:24>. The 3D model of AR SWS is created with variables height h, pitch px, py, top radius rxt, ryt and bottom radius rxb, ryb <ref:2610. p. 838165</ref:24>. This model represents the actual structures fabricated with ultra-short pulsed laser ablation <ref:2610. p. 738165</ref:49>. The condition (px/2) squared r 2xb + (py/2) squared ryb = 1 is used to derive the bottom radius in y <ref:2610. p. 838165</ref:49>.
Conversion between Linear Polarization
The polarization angle on the sky, α, is defined as an angle between the electric field vector on the sky and Xsky axis <ref:2610.38165</ref:27>. When a HWP works as shown in Equation (3), the detected power is given by 2S = Isky + P cos (4ρ − 4Φ + 4χ − 2ψ − 2α) <ref:2610. p. 638165</ref:27>. The matrix varies slightly from the ideal case and some of the elements result in M'QQ = MQQ cos 4(ρ − Φ) <ref:2610.38165</ref:27>.
Jones-Matrix Parameterization
The optical response is described using a Jones-matrix description, which allows HWP non-idealities to be described directly at the complex electric-field level <ref:2610.38165</ref:24>. The RCWA-computed response is expressed as an effective Jones matrix depending on frequency ν and HWP rotation angle ρ <ref:2610. p. 938165</ref:24>. The non-ideal part is expanded in Fourier modes of the HWP angle, δJα(ν, ρ) = X m=0,±2,±4 Am(ν)e imρ <ref:2610.38165</ref:24>.
By decomposing the Mueller coefficients into harmonics of the HWP rotation angle, we evaluated the HWP-synchronous components relevant to the detector signal <ref:2610.38165</ref:4>. The instrumental polarization is in the fourth-harmonic Mueller coefficients <ref:2610.38165</ref:4>. Its amplitude is typically of order 10−4 in the CMB bands <ref:2610.38165</ref:4>. This level is expected to be treatable with the mitigation method proposed by Patanchon et al. [22], provided that the HWP-induced signal is characterized and included in the time-ordered data model <ref:2610.38165</ref:24>. The corresponding additional polarization-angle rotation is approximately 30 arcsec at maximum <ref:2610.38165</ref:27>. The foreground spectral indices within representative ranges changes the band-averaged efficiency only at the percent level or below for most frequency bands <ref:2610.38165</ref:20>. The effect of the difference with the band coverage is also negligible if the instrumental polarization does not need to be modeled at better than the percent level <ref:2610.38165</ref:24>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>. By decomposing the Mueller coefficients into harmonics of the HWP rotation angle, we evaluated the HWP-synchronous components relevant to the detector signal <ref:2610.38165</ref:4>. The instrumental polarization is in the fourth-harmonic Mueller coefficients <ref:2610.38165</ref:4>. Its amplitude is typically of order 10−4 in the CMB bands <ref:2610.38165</ref:4>. This level is expected to be treatable with the mitigation method proposed by Patanchon et al. [22], provided that the HWP-induced signal is characterized and included in the time-ordered data model <ref:2610.38165</ref:24>. The corresponding additional polarization-angle rotation is approximately 30 arcsec at maximum <ref:2610.38165</ref:27>. The foreground spectral indices within representative ranges changes the band-averaged efficiency only at the percent level or below for most frequency bands <ref:2610.38165</ref:20>. The effect of the difference with the band coverage is also negligible if the instrumental polarization does not need to be modeled at better than the percent level <ref:2610.38165</ref:24>. The simple estimate of detector nonlinearity gives a negligible contribution for CMB anisotropy signals <ref:2610.38165</ref:23>.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Precise modeling of instrumental polarization requires using a Jones-matrix description to directly describe HWP non-idealities like differential transmission and phase retardance at the complex electric field level rather than relying solely on Mueller matrices which introduce ambiguities. This allows for a more direct characterization of instrumental polarization as
MIX and MXI (X = Q, U)
. -
Incorporate frequency dependence into the Jones matrix parameterization by expanding the non-ideal part in Fourier modes of the HWP angle, specifically including 2f and 4f components to capture residual rotation-angle dependence. This provides a
compact extension of the conventional frequency-dependent Jones parameterization to an effective response depending on both nu and rho
. -
Develop a simulation pipeline that integrates rigorous coupled-wave analysis (RCWA) with realistic sub-wavelength structures, such as those fabricated by ultra-short pulse laser ablation, to accurately model the
actual SWS geometry.
This allows for comparison between idealized models and fabricated structures to assess howhigher transmittance tends to be associated with lower instrumental polarization.
-
Implement a systematic error mitigation strategy that accounts for detector non-linearity by modeling the data model as d = S + cNLS 2 + O(S 3), and calculating harmonic-mixing terms based on the Mueller matrix decomposition. This addresses how
detector non-linearity can generate a spurious signal at 4frot through mixing of other harmonic components.
-
Enhance the analysis to include modeling of foreground spectral shapes, where the band-averaged efficiency is evaluated by varying spectral indices to assess sensitivity to spatial variations. This allows for quantifying how
foreground spectral indices within representative ranges changes the band-averaged efficiency only at the percent level or below for most frequency bands.
Related papers
- On-sky demonstration of a vector Zernike wavefront sensor in a cascaded adaptive optics system
- Euclid Data Release 1 (DR1): Calibration strategy and observations
- The General Antiparticle Spectrometer (GAPS) Antarctic Balloon Payload
- Detection and excision of radio-frequency interference in radio arrays
- Unified matched-filter and excess-power pipeline for gravitational-wave detector characterization
- Ushering in sub-arcsecond angular resolution observatories in the far-infrared: The case for a double-Fourier interferometer