Black hole winds at all scales

Proceedings IAU Symposium No. 378, 2023 © 2023 International Astronomical Union
A.C. Editor, B.D. Editor & C.E. Editor, eds. DOI: 00.0000/X000000000000000X

Searching for outflows in X-ray weak quasars

Bartolomeo Trefoloni1 ,2
1
Dipartimento di Fisica e Astronomia, Università di Firenze, via G. Sansone 1, 50019 Sesto
Fiorentino, Firenze, Italy
2
INAF – Osservatorio Astrofisico di Arcetri, Largo Enrico Fermi 5, I-50125 Firenze, Italy
email: [email protected]

Abstract. The connection between X-ray weakness and powerful X-ray outflows is both ex-
arXiv:2308.01345v1 [astro-ph.GA] 2 Aug 2023

pected in a scenario where outflows are connected with radiation-driven winds, and observed in
several sources, both in the local Universe and at high redshift. Here I present the first results
of a new study of this possible connection based on a search for SDSS quasars with weak X-ray
emission in serendipitous XMM-Newton observations. The selected objects have a ”normal” op-
tical/UV blue continuum, but a flat and extraordinarily weak X-ray spectrum. The availability
of rest-frame optical/UV spectra allows to check for the signature of outflows in the absorption
lines and/or in the profiles of the emission lines. This method could reveal the presence of a
population of so-far overlooked outflowing quasars and confirm the connection between winds
and X-ray weakness in quasars.
Keywords. quasar, AGN, outflows, X-rays, super-massive black hole

1. Introduction
Quasars are the most luminous persistent sources in the Universe, and there is grow-
ing evidence of their relevance for the observed properties of their host galaxies. It is
general thought that an accretion disc around super-massive black holes (BH) produces
their main contribution in the optical/UV (e.g., Salpeter 1964, Lynden-Bell 1969, Cz-
erny & Elvis 1987), while the X-ray emission is likely a product of the so-called “corona”
(e.g., Sunyaev & Titarchuck 1980, Haardt & Maraschi 1993), where UV disc photons are
boosted via inverse Compton scattering. These two emission features are closely related
(e.g. Lusso et al. 2020 and references therein), although the physics underlying such in-
terplay is still poorly understood, and their interplay could in principle vary with the
accretion parameters. At high accretion rate, for instance, the hypotheses of geometri-
cally thin and optically thick disc (Shakura & Sunyaev 1973) could break down, with the
thickening of the disc (Abramowicz et al. 1988, Chen & Wang 2004, Jiang et al. 2014).
In addition, powerful accretion-disc winds directly related to the nuclear activity (e.g.,
Proga 2005) could significantly alter the standard accretion process. Near Eddington
sources are more likely to host of such outflows (Zubovas & King 2013, Nardini et al.
2015), which could underlie the observed relations between the black hole mass (MBH )
and the galaxy properties (e.g., the MBH −σ relation; Ferrarese & Merrit 2000, Gebhardt
et al. 2000), although it is not yet clear whether and how AGN-driven outflows can affect
their host galaxies.
At the same time, several samples of highly accreting quasars sharing quite homoge-
neous UV properties have recently shown an enhanced fraction of objects whose X-ray
spectra are relatively flat and underluminous (by factors of > 3 − 10) with respect to the
expectations from the LX − LUV relation (e.g., Luo et al. 2015, Nardini et al. 2019, Zap-
1
2 Bartolomeo Trefoloni
pacosta et al. 2020, Laurenti et al. 2022). In many cases any clear evidence for absorption
has been revealed by the spectral analysis.

2. The parent sample

With the aim of expanding the LX − LUV relation sample in the high luminosity and
high redshift tail of the SDSS distribution, our group obtained dedicated XMM-Newton
observations of 30 luminous quasars between 3.0⩽ z ⩽3.3. This sample was selected in
order to present a high degree of homogeneity in terms of UV properties, while satisfying
the customary requirements of the LX − LUV relation sample (e.g. Risaliti & Lusso
2019). This implied the exclusion of known broad absorption lines (BALs) objects, as
well as Radio–loud (Fν,6cm /F > 10) and extincted (E(B − V ) >0.1) ones. Despite
ν,2500Å
sharing such optical/UV properties, the sample presented a diverse behaviour in terms
of X-ray characteristics (Nardini et al. 2019). About two thirds of the sample show X-
ray luminosities in agreement with the expectations from the LX − LUV relation (N
quasars), and a mean photon index of ΓX ∼ 1.85, fully consistent with typical quasars
at lower redshift, luminosity, and MBH (e.g. Bianchi et al. 2009). Their luminosity in
the 2–10 keV band spans between 4.5 × 1044 ⩽ L2−10 keV ⩽ 7.2 × 1045 erg/s, being one
of the most X-ray luminous samples of radio-quiet quasars ever observed. On the other
hand, one third of the sources proved underluminous by factors ∼ 3–10 (W quasars).
X-ray absorption at the redshift of the source is not generally statistically required by
the fits of the X-ray spectra. Notwithstanding the poor quality of the data in some of
cases, which leaves room for minor absorption, column densities NH (z) > 3 × 1022 cm−2
can be ruled out.
We continued our investigation about the spectral differences between N and W
quasars by performing a detailed analysis of the SDSS archival spectra for the whole
sample (Lusso et al. 2021). These spectra cover the rest-frame region roughly between
900-2200 Å including the C ivλ1549 emission line. We focused on such emission feature
to characterize its line properties (e.g., equivalent width, EW; line peak velocity, vpeak )
and underlying UV continuum slope as a function of the X-ray photon index and 2–
10 keV flux. We found that the composite spectrum of X-ray weak quasars is flatter
(αλ ∼ −0.6) than the one of X-ray normal quasars (αλ ∼ −1.5). The C iv emission line
is in general fainter in W quasars, but we did not report a strong blueshift (600–800
km/s) in both stacks. Such emission feature appears to be broader in the W composite
spectrum (FWHM ≃ 10,000 km/s) than in the N one (≃ 7,000 km/s), but this result is
not dissimilar from other literature samples at similar redshifts (e.g. Shen et al. 2011)
and luminosities (e.g. Vietri et al. 2018). We also added the sample from Timlin et al.
2020, adopting our selection criteria, with the aim to expand the dynamical range of the
parameters of interest. In such a way, we confirmed the statistically significant trends
of C iv vpeak and EW with UV luminosity at 2500 Å for both W and N , as well as
the correlation between X-ray weakness and the EW of C iv. In addition, a statistically
significant correlation between the hard X-ray flux and the integrated C iv flux for X-ray
normal quasars was found. This relation spans more than three decades in C iv and two
in terms of X-ray luminosity. X-ray weak quasars drop from the bulk of the relation by
more than 0.5 dex.
Lastly, our group was awarded new dedicated observations at the Large Binocular Tele-
scope (LBT) in the zJ and KS bands corresponding to the rest frame Fe iiUV -Mg iiλ2798
and the H β-[O iii] emission regions (Trefoloni et al. 2023). These new data allowed to
put tighter constraints on BH masses by taking advantage of H β and Mg iiλ2798 single-
 JD 11. Searching for outflows in X-ray weak quasars 3
epoch calibrations (Shen et al. 2011, Bongiorno et al. 2014), more reliable than the C iv
ones, so far available. We found that the sample is on average made of quasars accreting
near the Eddington limit < λEdd >= 1.2 (λEdd = Lbol /LEdd , being Lbol and LEdd re-
spectivelythe bolometric and the Eddington luminosity), as expected for such extremely
luminous objects. By fixing the slope of the continuum in order to match the one of the
bluer SDSS spectral counterpart, a trend has emerged with W objects generally exhibit-
ing more prominent Fe iiUV emission and, in turn, higher Fe ii/Mg ii ratios with respect
to N ones. Such feature ratio could be enhanced in case of wind-related microturbulence
and shocks in outflows (e.g. Baldwin et al. 2004, Sameshima et al. 2017, Temple et al.
2020). The EW [O iii] is generally low (<20 Å) in all but one of our objects. Inclination
effects are not expected to play a dominant role, since sources observed under a highly
inclined line of sight with respect to the axis of the disc usually display EW [O iii] ≳ 30 Å.
We also observed a lower EW of several emission lines (C iv, H β, [O iii]) in X-ray weak
quasars, and we argue that this effect could be related to the decrease of EUV photons
responsible for the line production. Indeed, the L[O iii] of the both X-ray weak and nor-
mal quasars is consistent with the high-luminosity extrapolation of the L[O iii] –L2−10 keV
relation from other samples in literature suggesting a common origin for the intrinsically
low X-ray and [O iii] emission, rather than an inclination/obscuration scenario.
We argue that, even though we do not diffusely observe unequivocal outflow footprints
(e.g. large C iv and/or [O iii] bluewings, BALs) in X-ray weak quasars, the presence
of a mostly equatorial disc wind could explain the observational features reported so
far (Figure 1). Part of the UV radiation would not be reprocessed in the X-ray corona
causing intrinsic (i.e., not due to absorption) X-ray weakness. The lack of distinctive
outflow signatures could be due to the modest inclination of the line of sight to the disc,
consistent with the observed EW [O iii] values that we find and with the huge bolometric
luminosities. A higher Fe iiUV emission in X-ray weak quasars could be produced by
regions of microturbulence and shocks at the interface between the wind and the BLR gas.
In addition, the effect of the change in the local mass accretion rate could dramatically
alter the EUV SED, and ultimately determine the observed difference in terms of emission
line strength (C ivλ1549, H β, [O iii]). The incidence of this mechanism could be higher
in samples of highly accreting sources.

3. A new pilot sample

Prompted by such results, we explored the possible correlation between the exotic coro-
nal state characterized by an underluminous and flat X-ray spectra and optical/UV out-
flow signatures. To this purpose, we designed a new pilot sample, made of less luminous
analogues of the W quasars, in order to investigate the tentative relation between the flat
and weak X-ray emission and the possible outflowing phase. We required the sample to
be X-ray underluminous by factors >3, while still presenting good X-ray and UV/optical
quality (SN≳5) in order to be able to detect absorption and/or outflows, if present. We
also required a photometric photon index Γ + δΓ < 1.3 (see Lusso et al. 2020 for more
details). Since for the optical/UV side of the emission we relied on archival SDSS data,
we restricted the analysis to the redshift intervals z⩽0.8 and 1.8⩽z⩽2.7 where respec-
tively the H β-[O iii]λλ4959,5007 complex and the C ivλ1549-Mg iiλ2798 fall. Adopting
such redshift intervals, we have also been able to retain at least one reliable BH mass
virial estimator to provide an estimate of the accretion parameters (MBH , λEdd ). These
criteria yielded a sample of 16 objects, 8 with UV spectra and 8 with optical spectra. We
performed a detailed spectral analysis on X-ray and optical/UV spectra. In the X-rays we
adopted a power law spectrum and included both the Galactic and the local absorption
4 Bartolomeo Trefoloni

X-RAY
[OIII]
NORMAL
2500 Å

CIV MgII Hβ
2 keV

TORUS
X-ray EUV Optical

[OIII]
X-RAY
WEAK
2500 Å

SHOCK
CIV MgII Hβ FeII
2 keV

DISC WIND
TORUS
X-ray EUV Optical
Figure 2. The LX − LU V sample from Lusso
et al. 2020, here used as a benchmark to select
Figure 1. A sketch describing the observa- X-ray weak sources. In red are depicted the W
tional features of X-ray normal (top) and weak sources. Large red stars represent the pilot sam-
(bottom) quasars. ple, squares the z∼3 sample.

with the last one turning out to be not statistically significant in general. This meant
that the observed weakness was intrinsic rather than due to obscuration. To perform
the spectral decomposition of the optical/UV spectra, we assumed a power-law contin-
uum and different emission line profiles (Gaussian, Lorentzian). We also allowed for a
blue component in both C ivλ1549 and [O iii]λλ4959,5007 and included templates for the
Fe iiUV and Fe iiopt pseudo-continua. Three sources out of eight with optical spectra dis-
play a prominent [O iii] bluewing with offset velocity in excess of ∼200 km/s and broad
≳850 km/s profiles, likely produced in galactic-scale outflows. Half of the sources with
UV spectra either show blueshifted (≳ 3000 km/s) or broad (≳2000 km/s) absorption
components in the C iv profile. As a rough comparison only 2% and 3% of the sources
reported in the latest SDSS catalogue (Wu & Shen 2022) exhibit blueshift in excess of
200 and 2000 km/s for the [O iii] and C iv emission lines respectively†. As a side note, we
should also mention that the X-ray and UV/optical spectral data have not been gath-
ered simultaneously, thus time shift effects could also dilute or even obliterate actual
correlations.
Although these preliminary results have been drawn from a relatively small pilot sam-
ple, a unifying picture relating the nuclear X-ray emission and the outer multi-scale
outflow phases is emerging (e.g. Zubovas & King 2014, Cicone et al. 2014, Gaspari et
al. 2020). In this framework, the weak phase of the corona could play a key role to set
the environmental conditions to launch fast winds. Otherwise an efficient X-ray emission
could overionize the nuclear medium, making wind line-driving unfeasible.
The results presented in these proceedings only represent a glimpse of what we believe
could bridge the observed state of X-ray underluminosity to the possibly related BLR
emission properties. Such multi-wavelength approach yields the potential to pave the way
for a more complete description and understanding of the interplay between these two
bands of the panchromatic quasar emission.
† Not being the parameters of the blue component available in such catalogue, these estimates
are based on the blueshift of the single broad component for the C iv and on the difference
between the global profile and the core peak wavelength for the [O iii].
 JD 11. Searching for outflows in X-ray weak quasars 5
References
Abramowicz, M. A., Czerny, B., Lasota, J. P., & Szuszkiewicz, E. 1988, ApJ, 332, 646-658.
Baldwin, J. A., Ferland, G. J., Korista, K. T., Hamann, F., LaCluyzé, A. 2004, ApJ, 615(2),
610.
Bianchi, S., Guainazzi, M., Matt, G., Bonilla, N. F., Ponti, G. 2009, A&A, 495(2), 421-430.
Bongiorno, A., Maiolino, R., Brusa, M., Marconi, A., Piconcelli, E., Lamastra, A., Cano-Dı́az,
M., Schulze, A., Magnelli, B., Vignali, C., Fiore, F., Menci, N., Cresci, G., La Franca, F.,
Merloni, A. 2014, MRAS, 443(3), 2077-2091.
Chen, L. H., & Wang, J. M. 2004. ApJ, 614(1), 101.
Cicone, C., Brusa, M., Ramos Almeida, C., Cresci, G., Husemann, B., Mainieri, V. 2018, Nature,
2(3), 176-178.
Czerny, B., & Elvis, M. 1987, ApJ, 321, 305-320.
Ferrarese, L., & Merritt, D. 2000, ApJ, 539(1), L9.
Gaspari, M., Tombesi, F., Cappi, M. 2020, Nature, 4(1), 10-13.
Gebhardt, K., Bender, R., Bower, G., Dressler, A., Faber, S. M., Filippenko, A.V., Green, R.,
Grillmair, C., Ho, L.C., Kormendy, J., Lauer, T.R., Magorrian, J., Pinkney, J., Richstone,
D., Tremaine, S. 2000, ApJ, 539(1), L13.
Haardt, F., & Maraschi, L. 1993, ApJ, 413, 507-517.
Jiang, Y. F., Blaes, O., Stone, J. M., Davis, S. W. 2019, ApJ, 885(2), 144.
Laurenti, M., Piconcelli, E., Zappacosta, L., Tombesi, F., Vignali, C., Bianchi, S., Marziani,
P., Vagnetti, F., Bongiorno, A., Bischetti, M., del Olmo, A., Lanzuisi, G., Luminari, A.,
Middei, R., Perri, M., Ricci, C., Vietri, G. 2022, A&A, 657, A57.
Luo, B., Brandt, W. N., Hall, P. B., Wu, Jianfeng, Anderson, S. F., Garmire, G. P., Gibson, R.
R., Plotkin, R. M., Richards, G. T., Schneider, D. P., Shemmer, O., Shen, Yue2015, The
Astrophysical Journal, 805(2), 122.
Lusso, E., Risaliti, G., Nardini, E., Bargiacchi, G., Benetti, M., Bisogni, S., Capozziello, S.,
Civano, F., Eggleston, L., Elvis, M., Fabbiano, G., Gilli, R., Marconi, A., Paolillo, M.,
Piedipalumbo, E., Salvestrini, F., Signorini, M., Vignali, C. 2020, A&A, 642, A150.
Lusso, E., Nardini, E., Bisogni, S., Risaliti, G., Gilli, R., Richards, G. T., Salvestrini, F., Vignali,
C., Bargiacchi, G., Civano, F., Elvis, M., Fabbiano, G., Marconi, A., Sacchi, A., Signorini,
M. 2021, A&A, 653, A158.
Lynden-Bell, D. 1969, Nature, 223, 690-694.
Nardini, E., Lusso, E., Risaliti, G., Bisogni, S., Civano, F., Elvis, M., Fabbiano, G., Gilli, R.,
Marconi, A., Salvestrini, F., Vignali, C. 2019, A&A, 632, A109.
Nardini, E., Reeves, J. N., Gofford, J., Harrison, F. A., Risaliti, G., Braito, V., Costa, M. T.,
Matzeu, G. A., Walton, D. J., Behar, E., Boggs, S. E., Christensen, F. E., Craig, W. W.,
Hailey, C. J., Matt, G., Miller, J. M., O’Brien, P. T., Stern, D., Turner, T. J., Ward, M. J.
2015, Science, 347(6224), 860-863.
Proga, D. 2005, ApJ, 630(1), L9.
Risaliti, G., & Lusso, E. 2019, Nature, 3(3), 272-277.
Salpeter, E. E. 1964, Publications of Goddard Space Flight Center, 165.
Sameshima, H., Yoshii, Y., Kawara, K. 2017, ApJ, 834(2), 203.
Shakura, N. I., & Sunyaev, R. A. 1973, A&A, 24, 337-355.
Shen, Yue, Richards, Gordon T., Strauss, Michael A., Hall, Patrick B., Schneider, Donald
P., Snedden, Stephanie, Bizyaev, Dmitry, Brewington, Howard, Malanushenko, Viktor,
Malanushenko, Elena, Oravetz, Dan, Pan, Kaike, Simmons, Audrey 2011, ApJS, 194(2),
45.
Sunyaev, R. A., & Titarchuk, L. G. 1980, A&A, 86, 121-138.
Temple, Matthew J., Ferland, Gary J., Rankine, Amy L., Hewett, Paul C., Badnell, N. R.,
Ballance, Connor P., Del Zanna, Giulio, Dufresne, Roger P. 2020, MNRAS, 496(3), 2565-
2576.
Timlin, J.D., Brandt, W.N., Ni, Q., Luo, B., Pu, Xingting, Schneider, D. P., Vivek, M., Yi, W.
2020, MNRAS, 492(1), 719-741.
Trefoloni, B., Lusso, E., Nardini, E., Risaliti, G., Bargiacchi, G., Bisogni, S., Civano, F.M.,
Elvis, M., Fabbiano, G.; Gilli, R., Marconi, A., Richards, G.T., Sacchi, A., Salvestrini, F.,
Signorini, M., Vignali, C. 2023, arXiv preprint arXiv:2305.07699.
6 Bartolomeo Trefoloni
Vietri, G., Piconcelli, E., Bischetti, M., Duras, F., Martocchia, S., Bongiorno, A., Marconi, A.,
Zappacosta, L., Bisogni, S., Bruni, G., Brusa, M., Comastri, A., Cresci, G., Feruglio, C.,
Giallongo, E., La Franca, F., Mainieri, V., Mannucci, F., Ricci, F., Sani, E., Testa, V.,
Tombesi, F., Vignali, C., Fiore, F. 2018, A&A, 617, A81.
Wang, J., Du, P., Hu, C., Netzer, H., Bai, J., Lu, K., Kaspi, S., Qiu, J., Li, Y., Wang, F.;
SEAMBH Collaboration. 2014, ApJ, 793(2), 108.
Wu, Q., & Shen, Y. 2022, ApJS, 263(2), 42.
Zappacosta, L., Piconcelli, E., Giustini, M., Vietri, G., Duras, F., Miniutti, G., Bischetti, M.,
Bongiorno, A., Brusa, M., Chiaberge, M., Comastri, A., Feruglio, C., Luminari, A., Marconi,
A., Ricci, C., Vignali, C., Fiore, F. 2020, A&A, 635, L5.
Zubovas, K., & King, A. 2013, ApJ, 769(1), 51.
Zubovas, K., & King, A. R. 2014, MNRAS, 439(1), 400-406.