Constraining dust physics with NIKA2 polarization : new road maps after Planck
The Planck mission has brought considerable constraints on the properties of dust, and their evolution from the most diffuse, high-latitude, Galactic ISM to the densest regions of the Gould Belt (Planck 2013 results. XI, Planck Int. Results 2015, XXIX). The combined analysis of ancillary and Planck HFI data has imposed a revision of dust models for the diffuse/translucent ISM, both in total and polarized emission (Planck Int. Results 2015, XXI, XXII). Planck maps have also revealed peaks of very high polarization fractions in the diffuse ISM (up to 20-22%, Planck Int. Results 2015 XIX, Planck 2018 results. XII), much higher than expected from models based on polarization observations in the optical (p/EBV = 9%), This high dust polarization efficiency has been recently confirmed by a direct follow-up in the optical by the RobolPol instrument, yielding p/EBV > 13% (Panopoulou+2019). The full-sky statistics of Planck polarization maps at 353GHz has allowed to demonstrate that the variations of the dust polarization fraction observed at 353GHz through the diffuse and translucent ISM were primarily driven by the magnetic field structure on the line of sight (Planck Int. Results 2015, XX), and not by a variation in the grain alignment efficiency, which was proved to remain high everywhere below a column density NH = 2.10^22 cm-2 (Planck 2018 results. XII).
At higher, subarcmin, resolution and higher optical depths, polarization measurements from starlight in the NIR, from dust emission in the FIR (SOFIA/HAWC+ at 89 and 154 µm) and submm (BLASTPol from 250 to 500 µm, POL-2 instrument on the JCMT/SCUBA-2 at 450 and 850 µm) allow to measure dust polarization at much higher extinctions (Av from a few tens to a few hundreds), with diverse conclusions. NIKA2 will probe Galactic dust polarization on scales of 0.01 to 0.1 pc, at resolutions comparable to those probed by SOFIA/HAWC+ and SCUBA-2/POL-2, but at longer wavelengths (1.3 mm/260GHz). Provided that we are able to disentangle in polarization observables between what is due to dust and what is due to the structure of the magnetic field on the line of sight, all these instruments allow to study how the grain polarization efficiency evolves with the density. In that perspective, I will present the recent methodology proposed by Planck results XII (2018). These studies are of particular importance to reassess the reliability of dust polarization as a tracer of the magnetic field in dense regions, but also for our understanding of dust evolution. Combining observations of the same regions at wavelengths ranging from 100 µm to 1 mm in total and polarized emission, we will also be able to study the correlation of the variation of the dust optical properties (variations in the dust emissivity and spectral index) with the observed variations in the dust polarization properties (alignment efficiency, shape).
Eventually, we should not exclude a priori the possibility of observing grain growth in high-density regions through the polarization expected from the self-scattering of dust emission by very large grains (Kataoka+2015). If such polarization signatures is detected, the joint analysis of FIR (SOFIA/HAWC), submm (SCUBA-2/POL-2) and mm (NIKA2) polarized observations may provide onstraints on grain size, as is now done for protoplanetary discs from multiple-bands ALMA observations (e.g. Yang+2019).