Probing frequency-dependent half-wave plate systematics for CMB experiments with full-sky beam convolution simulations

Abstract

We study systematic effects from half-wave plates (HWPs) for cosmic microwave background (CMB) experiments using full-sky time-domain beam convolution simulations. Using an optical model for a fiducial spaceborne two-lens refractor telescope, we investigate how different HWP configurations optimized for dichroic detectors centred at 95 and 150 GHz impact the reconstruction of primordial B-mode polarization. We pay particular attention to possible biases arising from the interaction of frequency dependent HWP non-idealities with polarized Galactic dust emission and the interaction between the HWP and the instrumental beam. To produce these simulations, we have extended the capabilities of the publicly available beamconv code. To our knowledge, we produce the first time-domain simulations that include both HWP non-idealities and realistic full-sky beam convolution. Our analysis shows how certain achromatic HWP configurations produce significant systematic polarization angle offsets that vary for sky components with different frequency dependence. Our analysis also demonstrates that once we account for interactions with HWPs, realistic beam models with non-negligible cross-polarization and sidelobes will cause significant B-mode residuals that will have to be extensively modelled in some cases.

Detailed study of HWP non-idealities and their impact on future measurements of CMB polarization anisotropies from space

“…Some complications could arise from the frequency dependence of those parameters, however. When this work was in preparation, a study by Duivenvoorden et al (2021) was published on the same topic, and we would like to point out that our analysis nicely complements the findings of those authors. Here, we offer a pedagogical approach to the use of a HWP in CMB experiments.…”

“…That is why our Mueller matrix elements depend only on the spinning angle of the HWP and does not depend on the incidence angle, which would be the case in general. To take into account non-orthogonal incidence, we should include the dependence on the angle of incidence in the computation of the Mueller matrix elements (Salatino et al 2018;Essinger-Hileman et al 2016) and convolve the total matrix by the beam (Duivenvoorden et al 2021). In this work, we have chosen a simplified approach neglecting the coupling between beam convolution and HWP non-idealities.…”

“…We refer to Patanchon et al (in prep. ); Salatino et al (2018), D'Alessandro et al (2019, Duivenvoorden et al (2021), Essinger-Hileman et al (2016 for studies that include the effect of slant incidence.…”

New physics from the polarised light of the cosmic microwave background

“…Miscalibration of polarisation-sensitive orientations of detectors on the focal plane of a telescope with respect to the sky as well as cross-polarisation coupling induced by systematics of beam [283][284][285][286] and optical components such as half-wave plates [287][288][289] yield α, hence a spurious B-mode power spectrum, C BB,o = sin 2 (2α)C EE . Reducing this to sufficiently small level required for the B-mode science sets a requirement for α, which is below 0.2 •290 or even below 0.1 •291 (see Fig.…”

Modelling ground pickup for microwave telescopes

“…Depending on the expected noise, the best filtering procedure might differ. In simulations and previous experiments, high-pass, pass-band, 32 or polynomial 33 filters have been applied to TOD. In our case, we implement a first-order Butterworth high-pass filter, with the cutoff frequency set to 0.1 Hz.…”