The physics of Active Galactic Nuclei
How much do we know about the UV polarization of Active Galactic Nuclei (AGN)? Only 4 AGN were observed with WUPPE at low polarimetric resolution (see the Mikulski Archive
for Space Telescopes). According to the Hubble Legacy Archive database, a total of 117 AGN were observed with different polarimetric instruments on board HST, out of which
only 35 in the UV with HST/FOS. However, the spectral resolution offered by the instrument was λ/∆λ ~ 1300 at best. The polarized UV band of AGN is thus an almost uncharted
territory. POLLUX will offer unique insight into the physics of AGN that is still little known, in particular by probing UV-emitting and absorbing material arising from
accretion disks, synchrotron emission in jet-dominated AGN and large-scale outflows.
At a parsec-scale distance from the accretion disk, an optically-thick dusty region probably forms along the equatorial region but is true geometry, size and composition remain unknown. A strong advantage for POLLUX is that the polarization induced by dust scattering rises rapidly toward the blue, peaking near 300 nm in the rest frame and remaining nearly constant at shorter wavelengths (see, e.g., Hines et al. 2001). Polarimetry at short wavelengths can thus discriminate between various grain models that provide different wavelength-dependent signatures and, if the grains are partially aligned, it can also help us to determine the magnetic fields topology and strength. Indeed, theory predicts that paramagnetic grains will be aligned with their longer axes perpendicular to the local magnetic field if exposed to magnetic fields (Lazarian & Hoang 2007). Therefore, POLLUX would not only selectively trace the smallest dust grains, allowing better characterization of AGN dust composition, but since the polarization degree is predicted to be proportional to the magnetic-field strength, polarimetry will also enable us to measure for the first time the intensity and direction of the magnetic field on parsec scales around the AGN core (Hoang et al. 2014).<
Finally, trapped between the accretion disk and the circumnuclear dust is the broad emission line region. This dense, rapidly moving (> 1000 km/s) medium is responsible for the broad lines observed in the UV, optical and IR spectrum of AGN and their detection in the polarized light of dust-obscured AGN lead to the bases of the unified model of AGN (Antonucci 1993). However, it was recently proven that a large fraction of non-detection of those broad lines in the scattering-induced polarized light was not due to a real absence of the region but rather to a non-sufficient spectral resolution. Thanks to the high-resolution capabilities of POLLUX, high signal-to-noise ratio spectropolarimetric evidence (or absence) of those broad lines can be revealed in a large sample of AGN, allowing in-depth tests of the unified model of AGN (Antonucci, Hurt & Miller 1994; Ramos Almeida et al. 2016).
In addition to jets, strong AGN outflows will be important targets for POLLUX. At redshift greater than 1.5 - 2, a sub-category of AGN, called Broad-Absorption-Line quasars (BAL QSO), exhibits very broad absorption features in UV resonant lines (Lyα, C IV, Si IV). BAL QSO are particularly interesting as they tend to have high polarization degrees (> 1%, e.g., Ogle et al. 1999), which can be used to constrain wind geometry (Young et al. 2007). These BAL QSO are believed to be the high-redshift analogues of more nearby, polar-scattered Seyfert galaxies, whose UV and optical emission can be explored by POLLUX. In particular, sensitive polarimetric observations would help investigate the dependence of broad absorption lines on bolometric luminosity and thus the role of radiative acceleration in the appearance of these lines (Arav & Li 1994; Arav et al. 1994).
Polarization can help to quantitatively determine the AGN contribution to feedback by measuring the synchrotron (de)polarization signatures (Mao et al. 2014). In addition, high-resolution UV spectropolarimetry can bring additional formation on the star forming processes thanks to the Zeeman effect associated with the intense local magnetic fields. The star forming period can be probed by POLLUX thanks to spectral line polarization, such as advocated for NGC 1808 using the Hα line (Scarrott et al. 1993). Extended surveys of emission line polarization compared to other AGN activity indicators across the visible spectrum could measure how important is the feedback in the evolutionary path of galaxies.
Probing the central AGN engine
Some key signatures of accretion disks can be revealed only in polarized light, and with higher contrast at ultraviolet than at longer wavelengths. Specifically, models of disk atmospheres usually assume Compton scattering in an electron-filled plasma, resulting in inclination-dependent polarization signatures (up to 10%, see e.g., Chandrasekhar 1960). Yet optical polarization is detected at less than a percent, and parallel to the radio jets if any (Stockman et al. 1979). Whether these low levels can be attributed to dominant absorption opacity (Laor & Netzer 1989) or complete Faraday depolarization (Agol & Blaes 1996) is unclear. This degeneracy can be broken by looking at the numerous UV lines that are formed in the innermost AGN regions (e.g., Lyα λ121.6, C II λ133.5, C IV λ154.9, Mg II λ280 ...). These lines are the key to understanding UV polarization, and only observations with high signal-to-noise ratio and the high spectral resolution of POLLUX can distinguish between the two effects.At a parsec-scale distance from the accretion disk, an optically-thick dusty region probably forms along the equatorial region but is true geometry, size and composition remain unknown. A strong advantage for POLLUX is that the polarization induced by dust scattering rises rapidly toward the blue, peaking near 300 nm in the rest frame and remaining nearly constant at shorter wavelengths (see, e.g., Hines et al. 2001). Polarimetry at short wavelengths can thus discriminate between various grain models that provide different wavelength-dependent signatures and, if the grains are partially aligned, it can also help us to determine the magnetic fields topology and strength. Indeed, theory predicts that paramagnetic grains will be aligned with their longer axes perpendicular to the local magnetic field if exposed to magnetic fields (Lazarian & Hoang 2007). Therefore, POLLUX would not only selectively trace the smallest dust grains, allowing better characterization of AGN dust composition, but since the polarization degree is predicted to be proportional to the magnetic-field strength, polarimetry will also enable us to measure for the first time the intensity and direction of the magnetic field on parsec scales around the AGN core (Hoang et al. 2014).<
Finally, trapped between the accretion disk and the circumnuclear dust is the broad emission line region. This dense, rapidly moving (> 1000 km/s) medium is responsible for the broad lines observed in the UV, optical and IR spectrum of AGN and their detection in the polarized light of dust-obscured AGN lead to the bases of the unified model of AGN (Antonucci 1993). However, it was recently proven that a large fraction of non-detection of those broad lines in the scattering-induced polarized light was not due to a real absence of the region but rather to a non-sufficient spectral resolution. Thanks to the high-resolution capabilities of POLLUX, high signal-to-noise ratio spectropolarimetric evidence (or absence) of those broad lines can be revealed in a large sample of AGN, allowing in-depth tests of the unified model of AGN (Antonucci, Hurt & Miller 1994; Ramos Almeida et al. 2016).
Outflows and jets
Magneto-rotational instabilities around the accretion disk are responsible for Poynting flux-dominated outflows (Blandford & Znajek 1977). The resulting jets tend to be collimated for a few parsecs and to dilute in giant lobes on kilo-parsec scales. Relativistic electrons traveling in ordered magnetic fields are responsible for the high polarization we detect (of the order of 40 - 60%, see e.g., Thomson et al. 1995). Interestingly, the continuum-polarization degree and angle are extremely sensitive to the strength and direction of the magnetic field, and to the charge distribution. By comparing the observed UV polarization of jetted AGN to leptonic, hadronic or alternative jet models, we will able to better constrain the composition and lifetimes of particles in the plasma. Since jets are also responsible for ion and neutrino emission, they are valuable sources to understand how cosmic rays are produced.In addition to jets, strong AGN outflows will be important targets for POLLUX. At redshift greater than 1.5 - 2, a sub-category of AGN, called Broad-Absorption-Line quasars (BAL QSO), exhibits very broad absorption features in UV resonant lines (Lyα, C IV, Si IV). BAL QSO are particularly interesting as they tend to have high polarization degrees (> 1%, e.g., Ogle et al. 1999), which can be used to constrain wind geometry (Young et al. 2007). These BAL QSO are believed to be the high-redshift analogues of more nearby, polar-scattered Seyfert galaxies, whose UV and optical emission can be explored by POLLUX. In particular, sensitive polarimetric observations would help investigate the dependence of broad absorption lines on bolometric luminosity and thus the role of radiative acceleration in the appearance of these lines (Arav & Li 1994; Arav et al. 1994).
Determining the impact of AGN onto galactic evolution
At even larger scales (0.5 arcsec/kpc at z = 0.1), UV polarimetric studies of young star-forming regions will allow to constrain the relation between the triggering of star formation and the onset of nuclear activity, which is still unknown (Hough 2006). The radiative and kinetic power transferred from the AGN to the host can easily quench, suppress or re-activate star-formation, profoundly altering the whole galaxy (Wagner et al. 2012).Polarization can help to quantitatively determine the AGN contribution to feedback by measuring the synchrotron (de)polarization signatures (Mao et al. 2014). In addition, high-resolution UV spectropolarimetry can bring additional formation on the star forming processes thanks to the Zeeman effect associated with the intense local magnetic fields. The star forming period can be probed by POLLUX thanks to spectral line polarization, such as advocated for NGC 1808 using the Hα line (Scarrott et al. 1993). Extended surveys of emission line polarization compared to other AGN activity indicators across the visible spectrum could measure how important is the feedback in the evolutionary path of galaxies.
 |
| Figure 1: Detectability of polarization in the GALEX/SDSS QSO catalog for a 15m (left) and an 8m LUVOIR aperture (right). The theoretical polarization degree in the u-band (322 +/- 26 nm) assumes a Gaussian distribution centered around 0.8% with a standard deviation of 0.3%, similar to what was observed in nearby Seyfert-1 galaxies. The observing time needed to reach a S/N of p/σp > 10 is color coded. About half the sample is beyond observation capabilities with an 8m aperture. |