MASARYKOVA
UNIVERZITA
Přírodovědecká fakulta
Ustav teoretické fyziky a astrofyziky
Systematický výzkum horkých atmosfér
obklopujících SO galaxie
Diplomová práce
Anna J uránová
Vedoucí práce: doc. Mgr. Norbert Werner, P h . D . Brno 2019
Bibliografický záznam
Autorka:
N á z e v práce:
S t u d i j n í program:
S t u d i j n í obor:
V e d o u c í p r á c e :
A k a d e m i c k ý rok:
P o č e t stran:
K l í č o v á slova:
Bc. A n n a Juráňová
Přírodovědecká fakulta, Masarykova univerzita
Ustav teoretické fyziky a astrofyziky
Systematický výzkum horkých atmosfér
obklopujících SO galaxie
Fyzika
Teoretická fyzika a astrofyzika
doc. Mgr. Norbert Werner, P h . D .
2018/2019
ix + 59
čočkovité galaxie, aktivní galaxie,
rentgenová spektroskopie, horký plyn
Bibliographie Entry
Author:
Title of Thesis:
Degree Programme:
Field of study:
Supervisor:
Academic Year:
Number of Pages:
Keywords:
Be. A n n a Jurahova
Faculty of Science, Masaryk University
Department of Theoretical Physics and
Astrophysics
A systematic study of the hot gaseous
atmospheres of SO galaxies
Physics
Theoretical Physics and Astrophysics
doc. Mgr. Norbert Werner, P h . D .
2018/2019
ix + 59
lenticular galaxies, active galaxies,
X-ray spectroscopy, hot gas
Abstrakt
Ř a d a masivních eliptických a čočkovitých galaxií, včetně těch s nezanedbatelným
momentem hybnosti, je prostoupena plynem o teplotách v řádu milionů kelvinů. Studie
procesů formujících termodynamický stav horkého plynu v nerotujících galaxiích jsou
publikovány již několik desítek let, avšak fyzikální vlastnosti horkých atmosfér těch
rotujících dosud nebyly systematicky analyzovány. V t é t o práci, zaměřené právě
na studium vlastností horkého plynu v rychle rotujících čočkovitých galaxiích, se
zabýváme analýzou pozorování rentgenové observatoře XMM-Newton. Z měření
vyplývá, že entropie je v centrálních oblastech atmosfér rotujících galaxií zvýšená
a její radiální n á r ů s t je pozvolný, což ve srovnání s nerotujícícmi objekty odpovídá
jejich efektivnějšímu ohřevu. N a základě změřeného systematicky nižšího tlaku plynu,
i při zohlednění celkové hmotnosti těchto galaxií, usuzujeme, že jejich horké atmosféry
mohou být expandující.
Abstract
Many massive elliptical and lenticular galaxies, including those with a significant net
angular momentum, are permeated by gas heated to millions of kelvins. Processes
shaping the thermodynamic state of the hot gas in non-rotating massive galaxies have
been studied over the last few decades. However, the physical properties of rotating
hot atmospheres and their connection to colder gas phases have not been investigated
systematically. Here, we present a study focused on properties of the hot gas in a sample
of fast-rotating lenticular galaxies, observed with the X-ray observatory XMM-Newton.
In comparison with slow-rotating systems, we measured an enhancement in central
entropy and overall flatter radial entropy profiles in rotating galaxies, indicative of
relatively higher efficiency of gas-heating in the central regions of these galaxies. We
found that the gas pressure in these objects is systemically lower, regardless of the
total galaxy mass, and conclude that the hot atmospheres of these objects could be
outflowing.
MASARYKOVA UNIVERZITA
Prírodovedecká fakulta
ZADÁNÍ DIPLOMOVÉ PRÁCE
Akademický rok: 2017/2018
Ustav: Ústav teoretické fyziky a astrofyziky
Studentka: Be. Anna J uranová
Program: Fyzika
Obor: Teoretická fyzika a astrofyzika
Směr: Astrofyzika
Ředitel Ústavu teoretické fyziky a astrofyziky PřF MU Vám ve smyslu Studijního a zkušebního řádu MU určuje diplomovou
práci s názvem:
Název práce anglicky: A systematic study of the hot gaseous atmospheres of SO galaxies
Oficiální zadání:
Massive early type galaxies are known to harbour extended atmospheres of hot X-ray emitting gas. A systematic study
of the thermodynamic properties of the hot X-ray emitting haloes has hitherto only been performed for samples of nonrotating
giant ellipticals. However, in the presence of significant angular momentum, heating from the central active
galactic nuclei (AGN) and radiative cooling are likely to proceed differently. The student will analyse X-ray data for a
sample of massive, rotating, SO galaxies, obtained with the XMM-Newton satellite (and possibly with the Chandra Xray
observatory) and determine the morphology and thermodynamic properties of the hot X-ray emitting atmospheres
surrounding these system. The student will compare the properties with those obtained for slow rotating giant elliptical
galaxies. The study will inform our understanding of the role of rotation in the formation of cooling instabilities in the
process of galaxy formation and evolution.
Jazyk závěrečné práce: angličtina
Vedoucí práce: doc. Mgr. Norbert Werner, Ph.D.
Datum zadání práce: 30. 11. 2017
V Brně dne: 16.1.2018
Souhlasím se zadáním (podpis, datum):
Název práce: Systematický výzkum horkých atmosfér obklopujících SO galaxie
Bc. Anna Juráňová
studentka
doc. Mgr. Norbert Werner, Ph.D.
vedoucí práce
prof. Rikard von Unge, Ph.D.
ředitel Ústavu teoretické fyziky a
astrofyziky
Poděkování
M y deepest gratitude belongs to my supervisor Norbert Werner for his sincere encouragement,
all the invaluable discussions, and patient guidance throughout this work.
Besides m y supervisor, I would also like to thank Filip Hroch for his generous help
and insightful comments, and K i r a n Lakhchaura for sharing with me her data. Many
thanks belong to my family for their endless support throughout my studies.
Prohlášení
Prohlašuji, že jsem svoji diplomovou práci vypracovala s a m o s t a t n ě s využitím informačních
zdrojů, které jsou v práci citovány.
Brno, 16. května 2019
Anna Juráňová
Contents
Introduction ix
1 Hot gas in galaxies 1
1.1 Formation of hot haloes 1
1.2 Influence of A G N 2
1.3 Thermal stability 3
1.4 W a r m and cold gas in early-type galaxies 5
1.5 Dynamical state of X-ray atmospheres 7
2 Studied sample 8
2.1 Selection criteria 8
2.2 N G C 3607 10
2.3 N G C 3665 10
2.4 N G C 4382 10
2.5 N G C 4459 H
2.6 N G C 4526 11
2.7 N G C 5353 H
2.8 N G C 1961 12
2.9 N G C 4649 12
3 Data analysis 14
3.1 Observations 14
3.2 Data reduction 15
3.3 Extraction of spectra 16
3.4 Spectral analysis 17
4 Results 19
4.1 Hot gas morphology 19
4.2 Thermodynamic properties 22
4.3 Thermal stability 32
5 Discussion 34
5.1 Hot gas morphology and mass 34
5.2 Thermodynamic properties 36
5.3 Thermal stability 39
vii
Contents viii
6 Summary and conclusions 41
A Supplementary material 42
Bibliography 47
Introduction
During the unending evolution of galaxies, some processes are inevitably hidden from
even the most inquisitive human eyes. Particles of gas expelled from stellar atmospheres
collide with each other, some of which heat up to form hot haloes that permeate
galaxies and even exceed their visible boundaries. In the other direction, from vast
and seemingly empty space, gas with a density lower than in ultra-high vacuum falls
into galactic gravitational potential wells and heats up to millions of kelvins to glow in
X-rays as well. In denser regions of space, the gas is brought also i n a more violent
manner - during mergers with other galaxies.
But not every galaxy can retain this X-ray emitting atmosphere - the gas can
diffuse out of the gravitational potential of the galaxy, it can be stripped off if the
host galaxy flies through surrounding medium or it can cool and thus leave energies
corresponding to emission of X-ray photons. The cold gas then serves as a source for
a formation of new stars or falls onto a central super-massive black hole, which is then
capable of launching jets and winds that heat and stir the surrounding medium. A
balance between all possible processes changes from galaxy to galaxy, which challenges
our understanding of not just the evolution of the hot phase, but galaxies and the
Universe whole.
It is well-known that spiral galaxies are rotating systems, but the same has been
recently measured i n many ellipticals and, of course, lenticulars, usually referred to
as a transitional type between the former two. Hot atmospheres of galaxies should
follow qualitatively same dynamical properties as their stellar components, hence in
fast rotating systems, X-ray haloes are subjected to a significant angular momentum.
This is expected to affect processes of cooling and heating, apart from being detectable
in morphological signatures of the overall motion.
Observations of hot haloes of fast-rotating galaxies are more complicated than those
of non-rotating giant ellipticals, which are due to entrapment in deeper potential wells
hotter and more luminous. However, the importance of understanding the processes
that shape these atmospheres and influence the evolution of their host galaxies lead
us to a study of the X-ray emission of rotationally supported galaxies in spite of the
practical obstacles.
ix
C H A P T E R 1
Hot gas in galaxies
Highly ionized gas confined within gravitational potential wells of galaxy clusters
has been successfully studied over last few decades and shed light on behaviour of
a supermassive black holes ( S M B H s ; e.g. Churazov et al., 2005) residing i n centres
of the brightest cluster galaxies ( B C G s ) , formation of these large-scale objects and
the chemical evolution of the Universe (e.g. Mernier et al., 2017). Finely tuned and
a rather gentle self-regulation process between the hot halo and an accreting S M B H
(active galactic nucleus, A G N ) , the so-called A G N feedback is becoming less mysterious
in these objects. However, in galaxies, where in principle this process is occurring too,
its efficiency i n retaining hot haloes for billions of years and preventing them from
cooling is not well understood (see e.g. Harrison, 2017). In this chapter, we summarize
the current understanding of hot atmospheres on galactic scales and expectations
resulting from simulations of these objects.
1.1 F o r m a t i o n of hot haloes
During the growth of galaxies on cosmological time-scales, hot gaseous atmospheres
were accreted externally and then augmented significantly by stellar mass loss. Today's
most massive slow rotating elliptical galaxies are believed to have formed i n quick
dissipative events lasting a few hundred million years, which produced massive compact
galaxies that subsequently grew further i n a series of 'dry' (cold gas free) mergers.
The fast rotating elliptical and lenticular (denoted SO) galaxies are believed to have
formed in 'wet' (gas rich) mergers of spiral galaxies (these merging systems are often
called Ultra-Luminous Infrared Galaxies - U L I R G S ) . Once the mass of the galaxy
exceeded approximately 1 O 1 2
M 0 , the galaxy became surrounded by a hot X-ray
emitting atmosphere.
The continuous deepening of the gravitational potential i n forming galaxies accelerated
clouds of extragalactic gas towards their centres. This externally gained
gas was heated to the virial temperature of the galaxy, T oc GM/R. The G here
is the gravitational constant, and M and R represent the galaxy mass and radius,
respectively. Young stellar populations of star-forming galaxies provided heating
and chemical enrichment v i a stellar winds and - mainly core-collapse - supernovae
(SNe), and as the star formation slowed down, heating via stellar explosions became
supplied by type l a SNe. A n imprint of these processes, translated to abundances of
1
1.2 Influence of A G N 2
heavier elements with transitions at energies in the X-ray band, should in principle be
measurable from spectra of the diffuse X-ray emission.
1.2 Influence of A G N
A n important role i n the evolution of the hot haloes and subsequently the whole
galaxies play the central S M B H s , which reside i n many massive galaxies and whose
past, as well as current, activity is often clearly visible i n X-ray observations. These
black holes have masses > 106
M 0 and processes connected to accretion onto them can
be broadly classified as either belonging to radiative or radio/mechanical feedback.
The former one, known also as a quasar mode, is mostly associated with high
luminosity A G N as most of the energy is released by radiation. The momentum
transfer from radiation to matter then creates outflows of velocity ~ 1 0 3
k m s _ 1
which
propagate through the galaxy, heating and lowering density of the hot medium and
destroying present dust (Morganti, 2017; Barnes et al., 2018; Cielo et al., 2018).
Strong radiative feedback requires a high accretion rate onto the S M B H ( M >
10~2
Medd)) which is not common in present day's ellipticals when radio-mode feedback
is dominant in suppressing cooling (Churazov et al., 2005). The Eddington accretion
limit is a m a x i m u m accretion rate defined for a black hole of mass M B H as M E C M =
47rGmP MBH/'(vc
<J
T)• The m p here is the proton mass, c speed of light, <TT Thomson
cross section, and r\ = 0.1 is the radiative efficiency of the accretion process. The limit
is given by a balance of the gravitational force and a force associated with radiation
pressure, acting in the opposite direction.
The most prominent evidence of radio-mode A G N activity are radio lobes and jets
emanating from galactic centres and reaching distances comparable to sizes of the
hosting galaxies or larger. The effect of these energetic processes on X-ray atmospheres
is directly observable in bubbles devoid of X-ray emitting gas, which reveal themselves
as 'cavities' of lowered surface brightness i n high-resolution X-ray images. These
cavities are coincident with radio lobes (where still present) and have been observed
in numerous sources, e.g. the B C G of the Perseus cluster (Boehringer et al., 1993)
or elliptical galaxy M 84 (Finoguenov and Jones, 2001). Processes involved in the
rise of radio bubbles have been found to trigger an uplift of gas from central regions,
redistributing colder and metal-rich gas to large distances from the central A G N
(Churazov et al., 2001; Simionescu et al., 2008). Similarly, nuclear outbursts are able
to produce shocks which propagate through the hot medium and deposit energy into
the atmospheres. They are observable as nearly azimuthally symmetric rings or arcs
of brightened X-ray emission (Forman et al., 2005). These spectacular manifestations
of A G N activity, which are capable of heating the X-ray atmospheres, lead us to a
question of the origin of material they 'feed on'.
The very fact that the hot gas radiates its energy away (and is thus observable)
implies cooling of the atmosphere as a whole. In the absence of heating, the atmosphere
would cool down and supply material for star formation i n a fraction of its age (e.g.
Fabian, 1994). Heating provided via A G N and SNe suppresses this catastrophic
scenario when the object - an individual galaxy or a galaxy cluster - would cool out
1.3 Thermal stability 3
of the X-ray band. However, at certain conditions, the atmosphere is still prone to
condensation and subsequently, cooling gas falling to the centre of the potential can
fuel the S M B H and trigger an episode of enhanced A G N activity. This, in turn, leads
to heating followed by a termination of the cooling phase.
It is critical to note that the feedback mechanism briefly outlined above is not
necessarily accurately sustained over large time-scales. This can, for example, happen
when the energy contained in jets is deposited beyond the X-ray atmosphere. Observational
evidence of this type of unbalanced cooling has been presented by e.g. Grossova
et al. (2019).
1.3 T h e r m a l stability
The question of the persistence of X-ray atmospheres has been addressed since the first
observations of thermal emission emanating from galaxy clusters and is still a major
subject of research in all its aspects. The time required for the hot gas to radiate away
its thermal energy, or shortly the cooling time, is here used in the form
where the electron and ion densities are denoted as n e and ri[, respectively, and
A(T) stands for the cooling function defined for a given temperature as an emission
weighted integral over energy, naturally depending on chemical composition (simplified
to metallicity Z) of the gas, which determines the contribution of atomic transitions
and overall properties of the gas emission. The form of equation (1.1) comes from
its broader definition, i.e. a ratio of gas internal energy, 3 / 2 ( n e + nCjk^T, and its
luminosity in unit volume nen\A{T^ Z), but it should be noted that also other definitions
are commonly used, employing gas enthalpy in the numerator of (1.1) instead (e.g.
Peterson and Fabian, 2006).
This quantity, measurable from spectral properties of the thermal X-ray emission,
served as a basis for the 'cooling flow' hypothesis (Fabian and Nulsen, 1977; Fabian,
1994). It is based on the assumption that the radiative energy losses of the gas are
compensated for by compressional heating as the cooling gas falls to the centre of the
gravitational potential. The flow itself would be provided through a decrease in i c o o ]
towards smaller radii, caused by an increase in density and a temperature drop. As a
result, the gas was expected to flow inwards while cooling, however this scenario has
been proven inapplicable to real objects, as expected effects across a wide range of
wavelengths are missing or much below theoretical predictions (Peterson et al., 2003).
Contrary to the lack of evidence for the uniform global cooling, observations of cold
gas, ongoing star formation, and A G N activity in galaxies and clusters require certain
amount of gas to cool anyway (e.g. O'Dea et al., 2008; Lakhchaura et al., 2018).
The importance of thermal instabilities in conditions present in hot atmospheres
was pointed out even before, by e.g. Field (1965). Thermally unstable gas is prone
to development of locally cooling clumps, and in the context of X-ray spectroscopy
would lead to formation of a multi-temperature spectrum. Conditions under which
the thermal instabilities can develop from perturbations in the hot plasma and grow
^cool
3 ( n e + n\)k^T
2 nemA(T, Z)
1.3 Thermal stability 4
non-linear depend on effectiveness of thermal conduction, viscosity and magnetic
fields (McCourt et al., 2011), and a prevalence of turbulent mixing (Scannapieco and
Brüggen, 2008) and heating (Zhuravleva et ah, 2014).
Rees and Ostriker (1977) suggested that the ratio of the cooling time (1.1) and the
gravitational free-fall time, defined for a given radius r and gravitational acceleration
g simply as
is a crucial quantity in determining whether the gas is thermally stable or not. They
argued that radiatively cooling pressure-supported clouds can re-adjust quasi-statically
and prevent a rapid cooling to temperatures of 1 0 4
K , as long as the cooling time
exceeds the free-fall time. In the opposite case, over-dense blobs, formed through local
non-linear perturbations in the hot medium, are expected to disconnect from the hot
phase and fall onto the A G N , as depicted in Pizzolato and Soker (2005), provided they
are prevented from evaporation by thermal conduction.
W i t h the advance of numerical simulations, this picture has been revisited and
modified by Sharma et al. (2012), who concluded that thermal instabilities develop
from linear perturbations and produce multiphase gas when the value of tC ooiAff fells
below approximately ten. Observationally, galaxies with ongoing cooling, and thus
having this ratio below the suggested threshold, are expected to produce detectable
emission from colder phases of atomic and molecular gas. Evidence for the validity
of this criterion has been presented for many systems, yet it does not seem to be
universally consistent with observations, as e.g. only 10 of 43 galaxy clusters studied in
Voit and Donahue (2015) with detectable H a emission have a minimum of measured
^cool/iff < 10, similarly in Hogan et al. (2017).
Generally, hot haloes do not seem to fall much below the tcoo\/ts « 1 0 threshold,
on scales of galaxies to galaxy clusters, as recently summarized i n Voit et al. (2018).
Combining previously published results for a number of objects (individual ellipticals,
groups and clusters), they concluded that processes operating i n hot atmospheres
are regulated so that they hold onto this ratio as a lower limit over seven orders of
magnitude in X-ray luminosity. Additionally, i n a sample of 96 early-type galaxies
(ellipticals or lenticulars, denoted E T G s ) , Babyk et al. (2018c) found no connection
of tcooi/tfi ratio and a presence or absence of observable cold gas. The cooling time
alone, on the contrary, seems to be closely connected to molecular gas content. Pulido
et al. (2018) and Babyk et al. (2018c) find a cooling time threshold of t c o o i < 1 G y r for
its presence in E T G s .
Striving to explain the processes of condensation from hot haloes, scenarios leading
to a formation of observed cold gas structures have been proposed over time. Numerical
simulations of cooling i n hot atmospheres, which predict sufficient feeding of the
central S M B H , have been performed by Gaspari and Churazov (2013). Subsonic
turbulent motions (originated in e.g. A G N activity or mergers, with velocity dispersion
av ~ 1 5 0 k m s _ 1
) i n their models significantly alter the process of condensation,
inducing chaotic gas dynamics, due to which perturbations become non-linear and
(1.2)
1.4 Warm and cold gas in early-type galaxies 5
form multiphase filaments detectable in longer wavelengths. This 'chaotic cold accretion'
( C C A ) and its direct connection to A G N heating should be capable of stabilising the
hot atmosphere at iC ooiAff ~ 10.
L i et al. (2015) predicted from their numerical simulations (omitting magnetic
fields and thermal conduction) that condensation should develop from gas uplifted by
AGN-triggered jets from its initial equilibrium position. The precipitation itself should
then proceed via interaction with the jet. Similar results i n terms of the influence of
jets were presented by Prasad et al. (2015), who used a different approach in numerical
simulations. T h i s 'stimulated feedback', i n which the precipitation from hot phase
is observed due to uplift from central regions, has been supported observationally in
several objects (McNamara et al., 2016; Tremblay et al., 2018).
The state of a hot atmosphere with respect to ongoing cooling has been recently
claimed to be given more accurately by a different dimensionless parameter than
tcooi/ts (Gaspari et al., 2018). Condensation via C C A , related to density-fluctuations
driven by turbulence, should be present in conditions where turbulent time-scale (£e ddy
defined below) is comparable to cooling time-scale, set by the proposed C-ratio
C = ^ E 2 L « l . (1.3)
teddy
The time required for a gyration of individual turbulent eddies, or the eddy turnover
time, is defined as
R 2 / 3 L L / 3
£eddy = 27T , (1.4)
where ov^ is the velocity dispersion at the injection scale length L , consistent with
a diameter of the cold/warm gas. The velocity dispersion is not easily measurable
even i n high-resolution X-ray spectra, but it can be obtained from measurements of
motions in products of the cooling process, the cold/warm gas.
1.4 W a r m a n d cold gas in early-type galaxies
Regardless of the exact mechanism, cooling from X-ray phase should naturally lead
to a presence of gas at lower temperatures, observable i n longer wavelengths. W a r m
(T ~ 1 0 4
K ) gas, bright in optical H a + f N n ] lines, is commonly observed in regions
associated with ongoing star formation, where the U V emission of young stars ionizes
the surrounding gas. However, the warm gas has also been observed i n galaxies and
galaxy clusters where the star-formation rate is not sufficient to warm the nearby
cold gas and the correlation between ionizing U V emission and H a has been found to
be weak and with a large scatter (McDonald et al., 2010). Supporting the evidence
for condensation from the hot phase, presence of H a filaments has been found to
strongly correlate with other properties suggesting a cooling of X-ray atmosphere (e.g.
McDonald et al., 2011).
These measurements are commonly performed with narrow-band filters or spectroscopically,
where N i l lines at 6548A and 6583 A are blended with the 6563 A H a line.
1.4 Warm and cold gas in early-type galaxies 6
In an analysis of 147 E T G s selected from volume-limited survey A T L A S for having
available data, 37% were detected in H a (Gavazzi et al., 2018).
Atomic hydrogen can be directly observed at wavelength A = 21.11 cm (or frequency
of v = 1.4 G H z ) thanks to the hyperfine splitting of the electronic ground state of
H i (Draine, 2011, pg. 70). Simulations of Lagos et al. (2014) predict that earlytype
galaxies acquire their atomic gas from radiative cooling of hot atmospheres
and, i n smaller amounts, through minor mergers or directly from the stellar mass
loss. Studying H i emission i n early-type galaxies, Lucero and Young (2013) found
comparable kinematic properties of atomic hydrogen and molecular gas, consistent
with a common origin of the two phases. Cold gas at temperatures T ~ 100 K can be
observable in far infra-red (FIR) spectra via forbidden lines [C n]A157 ixm, [Oi]A63|im
and [OiB]A145|im (Ferland et al., 1994). These emission lines have been observed in
e.g. filamentary emission nebula of multiphase gas likely uplifted by an A G N in M87
galaxy (Werner et al., 2013).
Even though the molecular hydrogen is the most abundant molecule in cold
molecular gas ( T < 5 0 K ) i n galaxies, it has no observable transitions equivalent to
temperatures that low and is therefore not directly observable. Instead, the second most
abundant molecule, C O , which has a permanent dipole moment, serves as a common
tracer of molecular gas. Observations of its transition between states with rotational
quantum number J = 1 to J = 0, C O ( J = 1 —> 0), at 2.6 m m have been established as
a useful tool for the total molecular hydrogen mass estimate, via C O luminosity to H2
mass conversion, in which N(H.2)/I(1 —>• 0) = 3 x 1 0 2 0
c m - 2
( K k m s - 1
) - 1
(Dickman
et ah, 1986). Transitions between higher rotational states C O ( J = 2 —> 1) at 1.3mm,
C O ( J = 3 —> 2) at 0.87mm trace gas at increasingly higher temperatures and densities
(e.g. Neininger et ah, 1998; Harris et ah, 2010). Cold molecular gas i n B C G s is most
probably a result of cooling from the hot intracluster medium, predicted from the
theoretical side and also observed near centres of clusters i n a form of thin filaments
with lengths on kiloparsec scale (Bridges and Irwin, 1998; Conselice et ah, 2001). In less
massive and smaller systems, groups and isolated galaxies, the issue is more complex,
allowing other processes to come into play. Recent surveys focusing on C O emission
in early-type galaxies, S A U R O N , A T L A S 3 D
and M A S S I V E found the presence of
molecular gas i n 28% (12/43 galaxies; Combes et ah, 2007), 2 2 % (56/259 galaxies;
Young et ah, 2011) and 25 % (17/67 galaxies; Davis et ah, 2019), respectively, of studied
objects. For the A T L A S 3 D
sample, Young et al. (2011) found no correlation with
stellar (K-band) luminosity, which implies that the molecular gas does not originate
purely in stellar mass loss. Total galaxy masses in their sample are not all above the
threshold for retaining hot haloes and therefore for those the origin i n condensation
from hot atmospheres can be ruled out. However, i n E T G s massive enough to host
hot atmospheres, Babyk et al. (2018c) found a correlation between mass X-ray halo
and molecular hydrogen content. The H2 was reported to be in systems with i c o o i as
short as 109
yr, indicating a condensation from hot atmospheres.
Apart from a gas of various properties, dust and polycyclic aromatic hydrocarbons
(PAHs) are observed in massive E T G s with cold gas in their centres, ranging from B C G s
to solitary galaxies (Donahue et ah, 2011; Kokusho et ah, 2017). P A H s are produced
in winds of evolved stars, just as the dust, and are generally easily destroyed by X-rays
1.5 Dynamical state of X-ray atmospheres 7
and hard U V photons. Hence, they are expected to be present in dense regions together
with dust and shielded by layers of X-ray absorbing medium (Monfredini et al., 2018).
1.5 D y n a m i c a l state of X - r a y atmospheres
In the general absence of strong cooling flows, hot atmospheres in isolated E T G s are
usually assumed to be i n or close to hydrostatic equilibrium. However, if the global
gas heating exceeds the radiative cooling, the atmosphere could change from stable
to out-flowing state. Pellegrini (2012) showed i n her work that the energy injection
through SNe at a given time, a function of stellar luminosity of the galaxy, becomes
dominant for low luminosity systems (with L B < 3 x 1 0 1 0
LB,© derived for properties
of early-type galaxies determined from usual scaling relations and for low redshift
galaxies, see Pellegrini, 2012, for more details). Although other processes influencing
the thermodynamic state of the hot gas operate in galaxies too, this threshold suggests
that outflows are likely present in low-mass systems and can have a significant effect
on properties of X-ray haloes. Furthermore, the dynamical state of the central region
can be decoupled from gas at larger radii, due to which only partial outflow (or inflow)
could be present.
Observations of Einstein observatory revealed that the X-ray luminosity L x of
early-type galaxies is related to its stellar shape. Derived ratios of L x / L B (thus
independent of stellar mass) for pure ellipticals were found to be systematically higher
than those of 'discy' ellipticals and SOs (Eskridge et al., 1995). The flattened shape
of the gravitational potential and rotational support, which has been confirmed for
flattened E T G s (Emsellem et al., 2011), should allow easier development of outflows.
From the view of thermal stability, Jurahova et al. (2019) found spectral features
indicating ongoing cooling i n the plane of rotation of a lenticular galaxy N G C 7049
which has a multiphase disc ranging from warm gas to cold molecular phase. The
spectra extracted from regions in the perpendicular direction, where the gas should be
supported only by buoyant force, were found to be consistent with single temperature
gas. These findings have been attributed to effects of rotation, where the thermally
unstable clumps of gas subjected to significant angular momentum are effectively
slowed down i n radial direction and therefore prevented from reaching equilibrium
position. This scenario has also been proposed from numerical simulations of Gaspari
et al. (2017), where the C C A in rotationally supported hot atmospheres should proceed
differently. In the case where rotation is the dominant motion i n the hot gas, i.e.
where the rotational velocity vTOt exceeds the velocity dispersion av (as in N G C 7049),
described by the so-called turbulent Taylor number T a t = vTOt/av > 1, the condensation
should proceed along helical paths and lead to a creation of a multiphase disc.
C H A P T E R 2
Studied sample
2.1 Selection criteria
To study the properties of the hot gas i n rotating lenticular galaxies, we selected our
objects based on several criteria. Given its origin, the gas motion can be probed
by that of the galaxy's stellar component, which we therefore used for a selection
of galaxies whose hot atmospheres are rotationally supported. The recent extensive
study of the kinematics of early-type galaxies, the A T L A S 3 D
Project (Cappellari et al.,
2011), provides an excellent basis for this selection, as among the main results are the
2D maps of stellar velocities measured along the line of sight.
Moreover, all the objects i n their sample have distances below 42 M p c , which
makes them satisfy a straightforward requirement on sufficient proximity, which would
allow us to perform a spatially resolved analysis of their extended X-ray emitting
atmospheres.
Taking into account the expected properties of the observed X-ray sources and
the main objectives of this work, we decided to use archival data from the X-ray
space observatory XMM-Newton (X-ray Multi-Mirror Mission, Jansen et al., 2001).
Compared to other X-ray telescopes with a spatial resolution needed in our case, the
instruments onboard of this spacecraft are sensitive i n energies below 1 keV. This
is a crucial quality for observations of SO galaxies, as the temperature of the gas
scales with the total mass of the galaxy, which is i n SOs generally lower than i n
giant ellipticals commonly observed in X-rays. This narrowed down the sample to six
galaxies, introduced in detail in the following sections, followed by two more objects: a
spiral galaxy with a hot halo - N G C 1961, and an elliptical galaxy N G C 4649, which
is, according to findings i n A T L A S 3 D
Project, also rotationally supported. Optical
images of the whole sample are shown in F i g . 2.1.
In the following sections, the studied objects are presented, focusing on properties
related to A G N and stellar feedback. Table 2.1 and Table 2.2 provide a quantitative
characterisation of properties relevant for this study, some of which were collected
using the N A S A / I P A C Extragalactic Database1
.
lr
The NASA/IPAC Extragalactic Database (NED) is operated by the Jet Propulsion Laboratory,
California Institute of Technology, under contract with the National Aeronautics and Space
Administration.
8
2.1 Selection criteria 9
NGC 1961
•
NGC 3607 NGC 3665
•
•
•
• •
•
•
NGC'4382 • NGC 4459
0.
NGC 4526
• •
—•
. • NGC 4649 NGC 5353
i
Figure 2.1: Optical images of the studied sample of SO galaxies, N G C 1961 and N G C 4649.
The images are taken from Sloan Digital Sky Survey (SDSS, data release 7) with an exception
of N G C 1961, which is from Digitalised Sky Survey 2. The solid line in the lower-left corner of
every image represents a scale of 1 arcmin.
2.2 N G C 3607 10
2.2 N G C 3607
N G C 3607 is an unbarred lenticular galaxy (morphological type SAO) and the brightest
member of Leo II group, and according to Giuricin et al. (2000) it is accompanied
by another 18 galaxies. It has the highest S F R i n our sample, 0.42 M 0 y r _ 1
. From
measurements of globular clusters kinematics, A l a b i et al. (2017) found that within
five effective radii, dark matter contributes to the total enclosed mass i n this galaxy
only by ~ 16 %, which is by a factor of 3 lower than i n other n o n - B C G s early-type
galaxies and what is predicted from cosmological hydrodynamical simulations. What
Alabi et al. propose as an explanation for this discrepancy is that the galaxy formed
much later on cosmological time-scales than most of the early-type galaxies. Chandra
X-ray Observatory measurements do not show evidence for an X-ray bright A G N .
2.3 N G C 3665
Another SAO type galaxy i n our sample, N G C 3665, is the brightest member of a
group of 11 galaxies (Makarov and Karachentsev, 2011). It has the highest luminosity
from the emission of polycyclic aromatic hydrocarbons, triggered by absorption of
far-ultraviolet light, for which it can be used as a tracer of ongoing star formation.
However, the star-formation rate measured by other methods is i n this system still
relatively low ~ 0.1 M 0 y r " 1
. N G C 3665 is among our radio brightest SOs at 1.4 G H z
and is the only one i n our sample that shows a pronounced radio activity associated
with an A G N . Twin jets were observed by Very Large Array (Parma et al., 1986), and
core radio emission obtained from very-large-array interferometry at 5 G H z (Liuzzo
et al., 2009). T h e mass of the supermassive black hole was measured by Onishi et al.
(2017) to be M . = 5.8 x 108
M 0 .
2.4 N G C 4382
This object, known also as M 85, resides i n the Virgo cluster and it is the only
galaxy i n our sample, which does not have a disc of cold gas near its centre. T h e
morphological type of this galaxy is SAO again, but this time, the galaxy also shows
shell-like structures. These features are believed to form from in-falling smaller galaxies
that oscillate i n the gravitational-potential well and form arcs of stars expanding to
the outskirts of the galaxy (see e.g. Thomson, 1991). The lack of cold gas is consistent
with observed low star formation, 0.002 M 0 y r _ 1
.
Capetti et al. (2009) found that the core of N G C 4382 does not show any radio
emission related to a central black hole, which is highly unusual for such a large earlytype
galaxy. More recently, Giiltekin et al. (2011) presented results from measurements
of stellar kinematics in the centre of N G C 4382 observed by Hubble Space Telescope,
which suggest that this galaxy may indeed not have a S M B H . More precisely, the
current black hole mass scaling relations, M , — L (Kormendy, 1993) and M , — a
(Ferrarese and Merritt, 2000; Gebhardt et al., 2000), predict a significantly larger
black hole mass when compared to their best-fitting value M , = 1.3 x 107
M 0 , a result
consistent with an anomalously low-mass black hole, if present. This result, along
2.5 N G C 4459 11
with no signs of past A G N activity, suggests that the observed X-ray halo could be
maintained purely by stellar feedback.
2.5 N G C 4459
Another member of the Virgo cluster, N G C 4459, is an unbarred lenticular galaxy
with a dusty disc extending out to r ~ 0.7 kpc, with observed blue clumps suggesting
a presence of newborn stars (Ferrarese et al., 2006). Unable to detect any neutral
hydrogen in their observations, Lucero and Young (2013) provide only an upper limit
on its mass, M m < 1.7 x 107
M 0 , despite the well measurable presence of molecular
hydrogen (see Table 2.2), yielding ratio M H 2 / M H I > 21. Far- and mid-infrared emission
of gas was studied i n detail by Young et al. (2009) and led to a detection of 24|im
emission from a disc reaching out to r ~ 2.6 kpc, which exceeds almost four times the
radius of the dusty structures and thus cannot be attributed to ongoing star formation.
They also find that N G C 4459 has the so-called FIR-excess, i.e. FIR-to-radio flux
density ratio exceeding a value of 3.04, a rare feature defined from observations of a
large sample of galaxies i n Y u n et al. (2001). In the centre of N G C 4459, Gavazzi et al.
(2018) report detection of H a emission. A n X-ray image from Chandra revealed a point
source co-spatial with the centre of the galaxy, and another point source separated by
~ 4.5arcsec from the first one.
2.6 N G C 4526
The only barred lenticular galaxy (SAB) i n our sample, N G C 4526, is also a member
of the Virgo Cluster. Having two occurrences i n the New General Catalogue, it is
also known as N G C 4560. Hereafter, we will only use the designation N G C 4526
for this object, which is also preferred i n the scientific community. It is oriented
nearly edge-on and also possesses a dusty disc i n the plane of rotation, spanning
over central r ~ 1.2 kpc. Star-formation rate is i n N G C 4526 the smallest among
galaxies with a dusty disc i n our sample and the P A H luminosity the second highest,
~ 17 x 1 0 4 1
W H z " 1
. A s i n the case of N G C 4459, Lucero and Young (2013) d i d not
detect neutral hydrogen and only put a n upper limit onto the M m < 1.9 x 1 0 7
M Q ,
while the total mass of present H2 was measured well (see Table 2.2). The lower limit
of their ratio is thus even larger than in N G C 4459: M H 2 / M H I > 100. Focusing on the
warm gas content, Gavazzi et al. (2018) found H a emission with a disc-like morphology
in this galaxy. Davis et al. (2013) measured the mass of the central supermassive black
hole to be M . = 4.5 x 108
M 0 .
2.7 N G C 5353
This almost edge-on oriented galaxy is a member of H C G 68 compact group with
N G C 5350 and N G C 5354 and two more objects (Hickson, 1982), accompanied by
other 45 fainter galaxies and more member candidates (Tully and Trentham, 2008).
It has an effective radius of only ~ 3.4 kpc, which is smaller than any of the other 5
2.8 N G C 1961 12
lenticular galaxies in this study. O'Sullivan et al. (2018) found a C O disc with a radius
of 0.8kpc and also a dusty disc in the central r ~ 0.5kpc was observed (Goullaud et al.,
2018). Its rotational velocity is very close to the velocity dispersion, which makes this
galaxy a great example of the transition between rotation-dominated galaxies and slow
rotators. This object is classified as a L I N E R (low-ionization nuclear emission-line
region) galaxy (e.g. Saikia et al., 2018). Sanchez-Sutil et al. (2006) reported an X-ray
bright nuclear point source emission spatially coincident with a radio source, which can
be associated with a central A G N . Finoguenov et al. (2007) studied the hot intragroup
medium of H C G 68, finding the temperature of the gas i n two radial bins from the
centre of the group: kBT(< 31 kpc) » 0.66 keV and kBT(< 92 kpc) » 0.69 keV with
overall metallicity measured only in the inner one, Z = (0.18 ± 0.04) ZQ.
2.8 N G C 1961
This rotation-dominated spiral galaxy is the brightest member of a small group of
seven galaxies (Tempel et al., 2016) and it has been classified as a L I N E R galaxy
(Carrillo et al., 1999). Measurements of rotational velocity at distances exceeding
10kpc from the galaxy centre (Rubin et al., 1979) have revealed that this spiral galaxy
is exceptionally massive, and at the same time, X-ray observations showing a point
source located i n the centre of the galaxy confirm a presence of an A G N . The total
mass of the galaxy is more than sufficient to retain its hot gaseous halo, which has
been observed by both Chandra and XMM-Newton. Bogdan et al. (2013) compared
available observations of N G C 1961 with results from numerical simulations finding
that the galaxy is dark matter dominated - baryons contribute to the total mass
with only 11 %. In radio observations at 6 and 18cm presented in Krips et al. (2007),
nuclear emission is accompanied by a ~ 2a signal resembling radio jets. Additionally,
distorted H i morphology revealed that the gas is being stripped by the surrounding
intragroup medium (Shostak et al., 1982).
2.9 N G C 4649
This elliptical galaxy, also known as M 60, resides in a group within the Virgo Cluster
(Mamon, 2008). Traces of motion of the galaxy with respect to the parent cluster
are pronounced i n the disturbed shape of its X-ray atmosphere, where ram-pressure
stripping and Kelvin-Helmholtz instabilities driven by the interaction of the hot
atmosphere of N G C 4649 and the intra-cluster medium were clearly observed (Wood
et al., 2017). Another deviation from the spherical symmetry, this time closer to the
galaxy centre, is connected to the A G N activity, as radio-jets inflated cavities i n the
X-ray gas can be observed in images from Chandra and are presented in e.g. Shurkin
et al. (2008); D u n n et al. (2010). This galaxy does not have clearly observable dusty
features, which is supported by no F I R emission detections reported by Temi et al.
(2003).
In A T L A S 3 D
Project, this galaxy is also classified as a fast rotator, but close to the
slow-/fast-rotator classification threshold. The support i n the X-ray gas atmosphere
2.9 N G C 4649 13
created by its rotation is negligible relative to the gas pressure support.
Table 2.1: Basic observational properties of the studied sample. Distance d is adopted from
A T L A S 3 D
Project and references therein with the exception of N G C 1961, in which case the
median value of distances from NED was taken, as well as values for redshift (z) in the second
column. The effective radii Re, velocity dispersions and rotational velocities are adopted from
Cappellari et al. (2011).
object d z scale Rc Vrot
N G C Mpc 10"3
arcsec k p c - 1
arcsec k m s - 1
k m s - 1
3607 22.2 3.14 9.29 38.9 222.0 ± 4 . 0 110.0 ± 9 . 0
3665 33.1 6.90 6.23 30.9 215.0 ± 8 . 5 94.3 ± 21.5
4382 17.9 2.43 11.52 66.1 68.3 ± 16.5 176.0 ± 3 . 5
4459 16.1 3.98 12.81 36.3 75.0 ± 20.0 171.8 ± 4 . 8
4526 16.4 2.06 12.58 44.7 150.4 ± 8 . 6 246.0 ± 6.0
5353 35.2 7.75 5.86 20.0 298.0 ± 9.0 284.0 ± 4 . 8
1961 32.4 13.12 6.37 51.0 242.0 ± 12.0 326.8 ± 9 . 5
4649 17.3 3.70 11.92 66.1 330.5 ± 4.6 55.0 ± 22.1
Table 2.2: Observational properties of the composition of the sample galaxies. B-band stellar
luminosity LB and B — V colour index are taken from HyperLEDA (Makarov et al., 2014),
1.4GHz radio power from Brown et al. (2011) and Condon et al. (2002), P A H luminosity from
Kokusho et al. (2017) and Stierwalt et al. (2014) in the case of N G C 1961, mass of molecular
hydrogen Young et al. (2011) and Combes et al. (2009) for NGC 1961, mass of atomic hydrogen
from Young et al. (2014) and Haan et al. (2008) for N G C 1961. P A H luminosity has not been
constrained for N G C 4649, just as atomic hydrogen content in N G C 4382 and N G C 4526.
object LB
B-V log P r a d i o £pAH l o g M H 2 l o g M H l
S F R
N G C 1 O 1 O
L B , 0 mag W H z " 1
104 1
e r g s - 1
M 0 M 0 M 0 y r " 1
3607 3.70 0.93 20.63 7.8 ± 6 . 2 8.42 < 6.53 0.420
3665 3.37 0.93 22.04 45.1 ± 9.9 8.91 < 7.05 0.109
4382 5.86 0.89 < 19.79 0.3 ± 5 . 5 < 7.39 - 0.002
4459 1.45 0.97 < 19.63 7.1 ± 2 . 9 8.24 < 6.53 0.071
4526 2.42 0.98 20.61 17.1 ± 4 . 4 8.59 — 0.028
5353 3.56 1.03 21.62 2.3 ± 6 . 7 < 7.44 < 7.07 0.095
1961 22.91 0.86 22.82 37.8 ± 3 . 1 10.39 10.67 9.24
4649 6.19 1.00 20.97 — < 7.44 < 7.19 0.129
C H A P T E R 3
Data analysis
3.1 Observations
Instruments which collected the data we use i n our analysis are named collectively
as European Photon Imaging Camera ( E P I C ) , each lying i n the focal plane of one
of the three X-ray telescopes onboard XMM-Newton X-ray observatory. Each of
these detectors covers a field of view 30 arcmin (Jansen et al., 2001) and has a useful
quantum efficiency at 0.2 — lOkeV.
The E P I C - p n camera consists of an array of 12 C C D chips which operate in parallel.
The observations we use were performed in a full frame or extended full frame operating
mode, with the active detector area covering the whole field of view. In the full frame
mode, each of the C C D subunits has an integration time of 68.7 ms followed by 4.6 ms
readout. Extended full frame mode has longer integration time, 199.2 ms, and the
same readout time (Striider et al., 2001). Photons detected during the readout period
(6.3 % in full frame and 2.3 % in extended full frame mode of the observation time) are
assigned to a wrong position and need to be accounted for properly during the data
analysis. The angular resolution of XMM-Newton is 15 arcsec half energy w i d t h at
1.5keV and thus much larger than the size of E P I C - p n pixels (4.1 arcsec side). The
energy resolution at 1.5 keV is ~ 110 eV (full width at half maximum, F W H M ) and
~ 150eV ( F W H M ) around 6keV (Striider et al., 2001).
Remaining two E P I C instruments, E P I C - M O S , are assembled of 7 M O S - t y p e
C C D s each, some of which have been lost over time due to micrometeorite hits. Only
44 % of the total flux reaches the detectors as half of the arriving photons are diverted
to reflection grating spectrometers. This allows a longer integration time, which is
2.6s and available continuously. One pixel covers only 1.1 x 1.1 arcsec and the energy
resolution at 1.5keV is also better than in E P I C - p n , ~ 80eV ( F W H M ) , but at 6keV,
the resolution decreases to ~ 150eV ( F W H M ) as well (Turner et al., 2001). Positions
of individual C C D s of all E P I C instruments are designed to overlap i n a way that
the detector chip gaps are mutually covered, in principle resulting in an image of the
whole field of view.
For the analysis described below, we used all E P I C observations with any useful
exposure time available at XMM-Newton Data Archive. These observations, denoted
by an observation ID ( O B S I D ) , are listed i n Table 3.1. In XMM-Newton E P I C
observations, each event (i.e. a detection on a C C D chip) is stored with information
on energy, detection time and position on the chip. This allows subsequent filtering of
14
3.2 Data reduction 15
Table 3.1: List of used observations for each galaxy and exposure times. In the second
column, observation ID is given, for which the total observation time ttot in MOS1, MOS2 and
pn detectors is shown in columns 3, 4, and 5, respectively, and the useful exposure time i n e t
(not contaminated by soft-proton flaring) is written in the next 3 columns. The sum of the
flaring-excluded exposure time in all instruments together is given in the last column.
object O B S I D
M l
itot [ks]
M 2 pn M l
inet [ks]
M 2 pn
E *net [ks]
N G C 1961 0673170101 31.5 31.5 30.4 21.5 23.5 11.3
0673170301 35.0 35.1 34.0 21.2 21.1 17.5
0723180101 21.8 21.8 20.8 19.8 19.6 15.9
0723180201 21.6 21.5 19.9 11.9 10.6 3.0
0723180301 23.9 23.8 22.7 16.3 18.1 9.3
0723180401 18.6 18.8 20.7 7.1 6.6 3.3
0723180601 25.6 25.5 23.9 9.1 9.8 7.0
0723180701 20.9 20.8 19.2 7.3 7.5 5.5
0723180801 14.7 14.6 19.0 13.6 13.6 9.7
0723180901 23.2 23.1 21.5 13.2 13.7 8.9 376.7
N G C 3607 0099030101 22.3 22.3 20.0 15.6 17.6 10.8
0693300101 44.4 44.4 43.8 31.3 35.3 19.8 130.4
N G C 3665 0052140201 40.6 40.6 36.3 27.8 29.6 20.3 77.6
N G C 4382 0201670101
0651910401
0651910501
0651910601
0651910701
33.5 33.5
41.7 41.6
41.2 41.3
41.2 41.1
36.5 36.5
33.2 18.8
40.6 33.6
40.2 29.9
40.1 27.7
34.9 28.3
18.8 14.4
34.1 25.7
29.6 24.1
28.7 22.1
28.3 24.2 388.3
N G C 4459 0550540101 81.8 81.8 82.8 72.7 73.2 60.6
0550540201 20.4 20.4 18.8 18.8 18.8 15.1 259.1
N G C 4526 0205010201 25.9 25.9 26.0 22.0 21.9 17.7 61.6
N G C 4649 0021540201 51.3 51.3 49.0 49.0 48.2 38.9
0502160101 81.7 81.7 80.5 72.9 72.7 61.0 342.6
N G C 5353 0041180401 22.3 22.3 20.0 21.3 21.1 16.8 59.1
events unrelated to the observation target, which is described below.
3.2 D a t a reduction
The data were reduced with standard procedures of XMM-Newton Science Analysis
System version 17.0.0. Event lists from raw-data files were obtained using tasks
emchain for E P I C - M O S and epchain in the case of EPIC-pn. Prior to filtering of the
output datasets, we also created a model of events received at times of relatively long
3.3 Extraction of spectra 16
readout of pn (the so-called out-of-time events). For this step, epchain task was run
for a second time with a corresponding option to obtain an observation-specific event
list (hereafter OOT-event list), which was, after proper scaling, subtracted from the
pn event list.
Almost every observation of XMM-Newton is contaminated by events triggered in
collisions of highly energetic protons with the detectors. These particles with energies
comparable to X-ray photons, focused by XMM-Newton optics and often referred to
as soft protons, originate in Solar activity and are trapped in Earth's magnetosphere
which is crossed by XMM-Newton during every revolution. In times of the highest
soft-proton (SP) flaring, the telescope is not operating to prevent permanent damage
to the detectors, but minor flares can be witnessed any time (Jansen et al., 2001).
The spectral profile of the S P component varies in time and is thus impossible to be
accounted for during the spectral analysis. Fortunately, periods of anomalously high
noise due to S P contamination can be defined and excluded from event lists based on
a light curve of individual observations. The SP unaffected times (good time intervals,
GTI) in our observations were determined based on a count-rate threshold set manually
from light curves observed at energies from 10 to 12keV, which is a commonly used
approach in this situation. The total and reduced exposure times resulting from the
G T I cleaning, £ t o t arid t n e t , respectively, are given in Table 3.1 to illustrate the quality
of the data used further on.
The event files were then filtered omitting detections nagged as unreliable, following
standard processing recommendations1
. W h e n an X-ray photon hits the detector, it
creates a geometrical pattern of pixels where any signal was detected. We excluded all
events unsatisfying a condition PATTERN<=12 for M O S 1 , M O S 2 which leaves out larger
than 'quadruple' patterns, and PATTERN<=4 for pn, resulting in a use of patterns quoted
only as 'single' or 'double'. Events excluded by these inequalities cannot be attributed
to genuine X - r a y photons and would pollute the observation if used. In addition to
these criteria, recommended strings #XMMEA_EM and #XMMEA_EP && (FLAG==0) were
used for filtering of E P I C - M O S and E P I C - p n event files, respectively, to apply further
corrections.
3.3 E x t r a c t i o n of spectra
To study the X-ray emitting gas in every galaxy presented in Chapter 2, we extracted
spectra from several concentric annuli, which allowed us to create radial profiles of
derived physical properties. O f course, this can be done only with the assumption of
spherical symmetry (azimuthal in projection), which will be discussed in the following
chapters. Where the number of received photons allowed it, we used the highest
reasonable spatial resolution, where the width of each annulus is given by the angular
resolution of XMM-Newton. This was unfortunately not the case of a majority of
galaxies in our sample, where the span of each annulus was determined by a number
of counts in a selection region that led to a reliable spectral analysis.
1
XMM-Newton Users Handbook, Issue 2.16, 2018 (ESA: XMM-Newton SOC, 2018)
3.4 Spectral analysis 17
Typically, X-ray point sources of various origin are projected onto the extended
emission of interest. To avoid their undesired contribution to the spectra, we encircled
the emission of each point source and excluded the selected events during the spectral
extraction procedure.
A s none of the X-ray atmospheres of interest covered the entire field of view,
it was possible to create a spectrum of local background to be subtracted prior to
spectral analysis. This approach to the handling of remaining events unrelated to the
examined source is based on the reasonable assumption that the background spectrum
is representative of the true background of the studied spectrum. Having the brightest
point sources removed by hand, the remaining contamination by distant A G N s does
not vary considerably. This also applies to diffuse soft X-ray foreground, which changes
at much larger spatial scales compared to the XMM-Newton field of view. Another
possible treatment is based on a modelling of all background components and thus
requires further assumptions on their spectral properties and, at the same time, the
sufficiently high number of received background photons to reliably constrain their
parameters.
3.4 Spectral analysis
A spectrum of a million-kelvin plasma i n our galaxies is determined primarily by
the temperature of the gas, its density and abundances of present elements. For
the spectral fitting, we used the S P E c t r a l X-ray and U V modelling and analysis
software ( S P E X , Kaastra et al., 1996) version 3.04.00, which uses an extensive atomic
database S P E X A C T (version 2.07.00). Apart from the classical Levenberg-Marquardt
minimisation of %2
algorithm, S P E X allows fitting using the C-statistic (Cash, 1979),
defined for Poissonian distribution of data. This is particularly useful for spectra
with a low number of counts i n a large set of bins where the distribution is far from
being Gaussian (and would thus require extensive binning which leads to a decrease in
energy resolution) and which is also the case of our data. Use of C-statistic is limited
to spectra with a positive number of counts per bin, hence they were binned to at
least one count per bin. We employed this statistic for the fitting of projected spectra,
using a model constructed in the following form.
A s the X-ray emitting gas can be described as a dilute plasma i n collisional
ionization equilibrium, we used the corresponding model cie in S P E X . A t the energy
resolution of E P I C instruments, the gas spectral properties are set mainly by the
temperature of the gas, its amount, which translates to a normalisation of the spectrum,
and elemental abundances. A s the latter were not possible to constrain properly, we
assumed Solar abundances of Lodders et al. (2009) and left free only the overall
metallicity of the gas. The fact that the gas temperature is not constant over the
volume of the atmosphere was accounted for by parameter sig. W h e n non-zero,
this parameter changes the model from single- to multi-temperature with a Gaussian
distribution and the root-mean-square width equal to sig. It is obvious that the
assumption of temperature distribution is not necessarily realistic, but still it serves
as a good approximation and adds only one free parameter to our model. Other
3.4 Spectral analysis 18
Table 3.2: Total hydrogen column densities taken from Kalberla et al. (2005).
N G C 1961 3607 3665 4382 4459 4526 4649 5353
i Y H , t o t [ 1 0 2 0
c m - 2
] 11.7 1.36 2.00 2.54 2.67 1.47 2.04 0.954
commonly used approaches are e.g. two cie components with one temperature being
tied to the second and by a factor of two smaller, or a different available component
with a given temperature distribution, neither of which would be better physically
motivated.
Low-mass X-ray binaries ( L M X B s ) present mainly in globular clusters were modelled
with one power-law component with spectral index 1.6, in accordance with Irwin
et al. (2003). Absorption by cold Galactic gas (at z = 0) was represented by a model
hot with the temperature set to 5 x 1 0 - 4
keV and column density of absorbing medium
taken from Leiden/Argentine/Bonn Survey (Kalberla et al., 2005, Table 3.2) and fixed.
Similarly, we assume the redshift of the source (Table 2.1) to be known precisely and
keep it fixed. The spectra were fitted in the energy range of 0.5 — 5.0keV, given that
the examined X-ray atmospheres produce most of the X-ray emission up to 2.0 keV,
while more energetic photons come mainly from a less steep decrease of the power-law
component describing the L M X B s . The lower limit on used energy is set by limitations
on proper calibration of incoming photons. From this range, events at 1.38 — 1.60 keV
were excluded where contamination from instrumental lines (e.g. Carter and Read,
2007) was present.
3.4.1 Deprojection
W i t h the hot gas being optically thin, the observed emission originates in the whole
volume of the atmosphere, which inevitably leads to effects related to projection i n
the observed spectra. Several possibilities for treatment of the data in such a situation
exist, all of which are based on the assumption of spherical symmetry of the extended
X-ray source.
The method adopted in this work is the so-called Direct X-ray Spectra Deprojection
( D S D E P R O J , Russell et al., 2008). A s the name suggests, spectra extracted from
individual annuli are, after scaling by area, subtracted from those lying closer to
their common centre. In order to prevent bins with a negative number of counts,
the spectra are binned to at least 25 counts per bin. Unfortunately, spectra created
with this method cannot be fitted using C-statistic, so %2
minimisation was used instead.
In the following chapters, results are given with la error bars and are derived for a
flat cold dark matter cosmology with HQ = 7 0 k m s _ 1
M p c - 1
and J7M = 0.3, J7A = 0.7.
Hereafter, the gas temperature is presented in units of equivalent energy, k^T.
C H A P T E R 4
Results
4.1 H o t gas m o r p h o l o g y
As will be discussed in the following section, the hot gas dominates the X-ray emission
in the energy range of 0.3 — 2.0keV. In order to examine morphological features and
verify our assumptions on spherical symmetry, we created images from the event files
including only the events corresponding to these energies. The final products, i.e.
background subtracted, exposure corrected and adaptively smoothed images developed
using tasks from Snowden and Kuntz (2011), are presented in F i g . 4.1. The signal is
displayed i n a logarithmic scale so that the hot atmospheres are visible out to their
outskirts while at the same time, bright point sources are removed in the same manner
as for the spectral analysis.
To determine the flattening of the atmospheres, we used the C I A O (version 4.11,
Fruscione et al., 2006) fitting tool Sherpa (Refsdal et al., 2009) to fit each of them
with a 2D /3-model (Cavaliere and Fusco-Femiano, 1976, 1978) in a form of
„1 -3/3/2
I(r) = I0 1 + r 2
, (4.1)
where
and
2 _ ( l - e x ) 2
* 2
+ ?/2
" r 2
( l - £ x ) 2 ( 4
- 2 )
x = (x - x0 )cos(9+ (y - yo)sin(9,
V = (y - Vo) cos 9 - (x - x0) sin 9,
with normalisation IQ, the (3 parameter, the centre of the emission [x,y], orientation
determining angle 9, and ellipticity ex left free. Parameters of interest, ex and position
angle P A x = 9 + 90° are listed in Table 4.1 along with the ellipticity and position
angle of stellar component taken from the literature. It is immediately obvious that
the position angles are generally very similar. The measured ellipticity is on average
lower i n the X-ray component but does not correlate directly with the ellipticity of
the stellar component. We note that some of the observed X-ray photons originated in
low-mass X-ray binaries and other stellar sources and we are therefore not probing the
hot gas alone.
19
4.1 Hot gas morphology 20
NGC 4649
W
Figure 4.1: Images of X-ray atmospheres of SO galaxies in our sample, N G C 1961 and
N G C 4649, all extracted in the energy range 0.3 — 2.0 keV. The images are displayed in
log-scale in order to visualize the full extent of the hot haloes, while the most prominent point
sources are removed. The solid line in the lower-left corner of every image represents a scale of
1 arcmin.
4.1 Hot gas morphology 21
Table 4.1: X-ray ellipticity ex and position angle (PAx) determined from /3-model fitting
and their optical counterparts, e+ and PA*, from Krajnovic et al. (2011) and Jarrett et al.
(2003) in the case of N G C 1961, for which the uncertainty has not been published.
object ex e* P A X [deg] P A * [deg]
N G C 3607 0.144 ± 0 . 0 1 1 0.13 ± 0 . 0 8 119.7 ± 2 . 3 124.8 ± 7 . 6
N G C 3665 0.158 ± 0 . 0 0 5 0.22 ± 0 . 0 1 28.2 ± 1.0 30.9 ± 2 . 0
N G C 4382 0.110 ± 0 . 0 0 6 0.25 ± 0 . 0 7 29.8 ± 1.7 12.3 ± 11.0
N G C 4459 0.060 ± 0.016 0.21 ± 0 . 0 3 134.1 ± 7.8 105.3 ± 1.9
N G C 4526 0.218 ± 0.005 0.76 ± 0.05 116.8 ± 0 . 7 113.7 ± 1.2
N G C 5353 0.253 ± 0.004 0.48 ± 0.04 136.6 ± 0 . 6 140.4 ± 4 . 9
N G C 1961 0.161 ± 0 . 0 0 9 0.330 100.8 ± 1.8 92.0 ± 2 . 0
N G C 4649 0.041 ± 0.002 0.16 ± 0 . 0 1 90.7 ± 1.6 91.3 ± 3 . 6
The influence of the surrounding medium on the X-ray emitting gas is clearly visible
in several objects. Observable deviation from radial symmetry i n N G C 3607 may
indicate ram-pressure stripping due to the motion of the galaxy within the intra-group
medium. The Leo II group has bi-modal X-ray brightness distribution (with the second
X-ray peak centred on N G C 3608 galaxy) and according to Mulchaey et al. (2003) the
group members could be in a process of merging.
Ram-pressure stripping is more pronounced in N G C 4459, where an X-ray emitting
tail formed and visualises the direction of tangential motion of the galaxy within the
Virgo cluster. A n X-ray source on the right from N G C 4459 in F i g . 4.1 has been
excluded from the spectral analysis. Based on available images from optical telescopes,
it cannot be attributed to a nearby galaxy cluster, where individual galaxies should be
clearly observable in optical images. We discuss the possible nature of this source in
Chapter 5.
In the case of N G C 5353, the position of the centre of the galaxy lies in mutually
overlapping chip gaps and bad pixels and therefore there is the black spot in the centre
of the X-ray image as no emission could be detected. The point source removed in the
upper part of the extended emission can be geometrically attributed to an A G N i n
N G C 5354 and the surrounding diffuse emission probably forms its hot atmosphere.
On the opposite side from this source, another enhancement in surface brightness was
detected, but i n this case, an optical counterpart cannot explain its presence. A s in
the previous case, this region has been also excluded from spectral analysis.
As was already mentioned in the previous chapter, N G C 4649 is undergoing rampressure
stripping and the effects of this process are observable in the outskirts of the
detected X-ray emission.
In the rest of this chapter, the results are derived from spectra extracted from
concentric annular regions and the error bars on radius represent the width of these
annuli.
4.2 Thermodynamic properties 22
4.2 T h e r m o d y n a m i c properties
Due to the limited energy resolution of E P I C instruments, low-count spectra of
diffuse X-ray emitting plasma from galaxies show strong anti-correlation between gas
metallicity and normalisation, as models with enhanced metallicity are statistically
indistinguishable from models with lower metallicity but increased normalisation (see
Fig. 4.2 for an illustration). The resulting degeneracy can be removed or at least
reduced by increasing the number of counts. A s no other observations were available,
the photon count increase could only be obtained by enlarging the geometric region for
the spectrum extraction. This ultimately leads to a combination of photons emitted
from the gas of different properties and thus a formation of a multicomponent spectrum
of the gas. This, unfortunately, further complicates the situation and does not result
in any improvement in metallicity determination.
Energy [keV]
Figure 4.2: Background subtracted spectrum of X-ray emission from the second radial bin
of N G C 5353 from all three instruments. Orange, red, and blue lines are best-fitting models
composed of redshifted and absorbed cie and a power law component, where normalisations
for the latter two components are allowed to vary, just as the cie temperature. All three
provide a good fit to the data. Residuals below the spectrum are black for the metallicity used
throughout, Z = 0.5 Z Q , and orange and blue for 0.25 Z Q and 1.0ZQ , respectively. The fit was
performed on unbinned data. Due to a larger effective area, the spectrum of pn instrument
lies above those of MOS1 and MOS2.
For almost all spectra in our analysis suffer from this degeneracy, we fixed the
overall metallicity of the gas to 0.5 Z Q , which is a reasonable estimate based on the
origin of the gas (stellar mass loss and accretion of intergalactic medium with a low
4.2 Thermodynamic properties 23
Table 4.2: Total X-ray gas luminosity, emission-weighted temperature obtained from global
spectra, and mass. Where a single temperature model did not provide a good fit, a multitemperature
model was used and the best-fitting value of the additional parameter CTTx is
presented.
object Lx kBTx
1 0 4 0
e r g s " 1
keV keV 109
M 0
N G C 3607 1.74 n 4 1 1 +0.025u
- 4 i i
- 0 . 0 0 9
•, ^9+0.07
N G C 3665 2.30 n oi 9+O.OO6 1 nq+0
-0 7
i
-u y
-0.07
N G C 4382 7.97 n oifi+0.025
U.O±O_0.024 n i 2 + a 0 1 7
u . u i z _ 0 0 1 2
5 2 6 + a l
°°-Z D
-0.10
N G C 4459 0.31 o i q n + 0
- 0 4 1
- 0 1 2 + a 0 2
u
- i z
- 0 . 0 2
N G C 4526 0.71 0 260+0
-0 1 3
U . Z D U _ 0 0 1 9
- 0 1 5 + a
° 3
u
- i o
- 0 . 0 2
N G C 5353 4.21 0-65118^8 N 99^+0.042
u.zzo_0 0 4 6
n 4Q+0
-0 3
u
-4 y
-0.03
N G C 1961 4.79 0 . 2 9 8 + ^ 0 202+0
-0 8 2
6 01 + ° - 4 9
° - u i
- 0 . 4 6
N G C 4649 68.26 0.8791°;°°} - i u
- '°-0.61
level of chemical enrichment) and is commonly used i n such situations. The only
exception is N G C 4649, for which the metallicity was reliably determined from the fit
of the global spectrum, Z = (0.73 ± 0.04) Z Q , i n accordance with results of Mernier
et al. (2017). For the deprojection analysis, the metallicity of this object was set to
0.7 Z 0 and fixed.
4.2.1 Global properties
We estimated the total gas mass Mx of the hot atmosphere by summing total particle
number density n derived from deprojected spectra in a given spherical shell, i.e.
Mx = 47T ^2 r2
p(r)Ar = iirmnH X ! r 2 n
( r
) A r
> (4
-4
)
where \x is mean atomic weight, \x = 0.62, and mu is the mass of a hydrogen atom.
The particle number density was calculated as n = 1.92 n e , where the electron number
density n e was obtained directly from the spectrum normalisation. The summation
has been performed out to the last annulus presented i n this work, corresponding to
approximately 2 — 6 Re, set by the data quality.
The calculated masses are given in Table 4.2. For the density profiles, see Fig. A . 2
in the Appendix. We fitted the entire galaxy X-ray gas emission with a single- or
multi-temperature model, depending on the spectrum (see sec. 3.4), to obtain global
properties of the hot gas. Best-fitting parameters derived from this fit are listed i n
Table 4.2. Gaussian width of the temperature distribution of multi-temperature model
is denoted as CTTx and is presented for spectra where this model provided a better fit.
4.2 Thermodynamic properties 24
1.0-
0.8-
0.6
0.4
0.2 JGC 1961 NGC 3607
i NGC 3665
NGC 4382
i r
NGC 4459
i r
NGC 4526
NGC 4649
—
r -
4
NGC 5353
- r -
4
-i r -
0 2
r [Reff]
Figure 4.3: Radial, azimuthally averaged profiles of deprojected temperature with metallicity
fixed at 0.5 Z Q except for N G C 4649 for which the metallicity is 0 . 7 Z 0 . For clarity, dark blue
points represent SO galaxies in our sample, while the spiral galaxy N G C 1961 and the elliptical
N G C 4649 are plotted in light blue.
4.2.2 Temperature
The best-fitting temperature obtained directly from both projected and deprojected
spectra of every annulus is presented i n F i g . 4.3 as a function of effective radius, and
thus scaled by the radial extent of the stellar component. Temperature profiles i n
physical units derived only from deprojected spectra, which we used for the analysis
further on, can be found i n the Appendix, Fig. A . l . The spiral and elliptical galaxy are
plotted i n a lighter shade of blue to be easily distinguishable from lenticulars. The most
striking feature of the mosaic is that the mean temperature of SO galaxies is smaller
than that of the only elliptical (and more massive) galaxy N G C 4649. Overall, the
profiles of SO galaxies do not show any significant trends and within 3a uncertainties
are close to being isothermal.
4.2 Thermodynamic properties 25
10"
10 -11 .
10"
10"
S IO-1 1
O l
a 1 0 _ i 2
10"
10"
10"
NGC 1961
NGC 4382
NGC 4649
™ i —
10
NGC 3607
NGC 4459
+
NGC 5353
1 10
r [kpc]
NGC 3665
NGC 4526
10
Figure 4.4: Pressure profiles derived from deprojected spectra. Point colour is chosen as in
Fig. 4.3 and is kept in remaining figures of this kind further on.
4.2.3 Pressure
The gas pressure as a function of the radius was calculated assuming ideal gas, i.e.
p = nkBT. (4.5)
These profiles are shown separately in F i g . 4.4, using the same colour-coding as for the
temperature above. In all objects, the pressure is monotonically radially decreasing,
which is expected for gravitationally stratified atmospheres. T h e main difference
in profiles of rotating galaxies i n our sample (SOs and N G C 1961) and the massive
elliptical is the vertical shift within the plot, or i n other words, the pressure is almost
an order of magnitude higher i n N G C 4649 with respect to the remaining seven objects.
W h e n compared to a larger number of elliptical galaxies, namely 49 ellipticals
(including N G C 4649) studied by Lakhchaura et al. (2018), the division remains clearly
visible, as can be seen i n F i g . 4.5. Their profiles are distinguished by a presence and
4.2 Thermodynamic properties 26
10 -9 _
10 -10 _
E
u
at
ai 1 0 " 1 1
10 -12
- i — I — I — r - p -i—I—I—r-| 1 1 1 1—I—I—I—r-p
Cool Gas Free
Nuclear Cool Gas
Extended Cool Gas
NGC 1961
NGC 3607
NGC 3665
NGC 4382
NGC 4459
NGC 4526
NGC 4649
NGC 5353
_l I I I I l_
10 100
r[kpc]
Figure 4.5: Pressure profiles of all eight galaxies (lines) and median profiles of ellipticals
(Lakhchaura et al., 2018) distinguished by cool gas content. The shaded regions represent a
median absolute deviation spread.
morphology of 'cool' (relative to X-ray) gas, more specifically of H a + [ N n ] emission.
The solid red, dot-dashed green, and dashed blue lines stand for median entropy
profiles of galaxies undetected i n H a + f N n ] , with nuclear emission and having cool
gas with filamentary structure, respectively. The surrounding shaded regions represent
median absolute deviation ( M A D ) . Regardless of the cool gas content in their sample
(further discussed i n the text below), the pressure in SOs is lower at all radii. One
of the reasons for this difference lies in the total mass of these galaxies. Ellipticals
are known to be generally more massive than lenticular galaxies and the difference in
gravitational potential should also reflect i n the gas pressure. Rotation can further
lower the effective potential, allowing the gas to remain more dilute.
To limit the effect of different total mass, the radial distances need to be plotted
in properly scaled units. The determination of the total enclosed mass M ( < r) within
some radius r is computable from the thermodynamic properties of the gas, i.e.
dp GM{< r)p
-r- = 2 ' (4
-6
)
dr r z
assuming spherically symmetric atmosphere in hydrostatic equilibrium and a negligible
effect of non-thermal pressure. This, after a reorganisation and a substitution for the
4.2 Thermodynamic properties 27
gas density, p = mnpn, yields
M(<r) = - - ^ — ^ . (4.7)
Gmnpn ar
Mass-scaled radius is then usually defined through the mean enclosed density in units
of the critical density of the Universe,
Pc = ^ , (4.8)
where we assume the Hubble constant Ho = 7 0 k m s _ 1
M p c - 1
. The scale radius r<\,
representing the distance from the galaxy centre at which the mean enclosed density
is equal to A p c , is given simply through
, a = A p c = ^ ) . ( 4 . 9 )
After substitution with (4.7) and several simplifications, it can be written as
1 1 dp
map r^n{r/±) dr
(4.10)
''A
For a direct comparison with the scale radius obtained differently, we set A = 200.
Solving numerically this equation, we obtained values of r2oo listed in Table 4.3. For
the pressure gradient, we use a functional form i n the calculation. The pressure can
be well described by /^-profile (Cavaliere and Fusco-Femiano, 1976, 1978), defined as
p(r) = po 1 +
r ,2
-3/3/2
(4.11)
where the core radius r c , central pressure po and parameter (5 are determined from
the fit. In the case of N G C 3607, where gas properties have been determined for only
three radial distances, a power-law function was assumed instead. Particle density n
is also expressed as a function of radius, assuming a /3-profile or a power law again.
However, the requirement of hydrostatic equilibrium is obviously not fulfilled i n
rotating systems, and therefore the obtained values should serve only as a rough
estimate. The M200 for five galaxies in our sample, among others, has been determined
by James et ah (2018). In that work, an independent method has been used, employing
the kinematics of globular clusters to determine the total mass of the galaxy, or using
the total stellar mass of globular clusters as a proxy for the total gas of the galaxy,
as the two have been found to correlate. The values of r2oo,GC derived from these
measurements are listed also in Table 4.3. For N G C 1961 we adopt r2oo determined
in Bogdan et ah (2013) from baryonic Tully-Fisher relation for the cold dark matter
cosmogony, M200 oc V ^ 3
, where F m a x is the maximum rotational velocity of the galaxy.
Even though likely more accurate than the one derived from X-ray gas properties, this
value should be regarded also as an estimate. Plotting the pressure profiles of SOs
against r2oo along with ellipticals (Fig. 4.6), the dichotomy is still clearly visible. The
4.2 Thermodynamic properties 28
Table 4.3: Scale radius T2oo, x, derived from X-ray observations, and T2oo, GCi which has been
determined from globular clusters kinematics or the total mass of galaxy's globular clusters
from James et al. (2018). *The value has been determined from maximal stellar rotation
velocity in Bogdan et al. (2013).
N G C 1961 3607 3665 4382 4459 4526 4649 5353
T200,X [kpc]
^200, GC [kpc]
213.1
470*
410.4
396.1
272.0 295.7
491.0
677.7
345.0
236.7
402.2
730.0
749.0
829.5
dashed grey line represents a median profile of pressure in 19 ellipticals of Lakhchaura
et al. (2018), for which the r2oo,GC is also known from the globular clusters properties,
and the grey-shaded region shows the corresponding median absolute deviation. SOs
are further distinguished by a rotational velocity, showing no observable systemic
trend.
10"
E
u
a 1 0 - u
10 •12
• Ellipticals
NGC 3607
NGC 4382
NGC 4459
NGC 4526
0.001 0.01
NGC 4382 -
0.1
246.0
[ 176.0
" [ 171.8
110.0
r [r2 0 0 ]
Figure 4.6: Deprojected pressure profiles of SOs with known r2oo,GC> displayed in the scale
radius which was obtained by a method independent of dynamical properties of the galaxies.
The dashed grey line stands for a median profile of elliptical galaxies studied in Lakhchaura
et al. (2018) and the surrounding grey region defines the median absolute deviation spread.
The rotational velocity of the galaxy is represented by the line colour.
4.2 Thermodynamic properties 29
100
100
N
E
100
NGC 1961
1—
NGC 3607 NGC 3665
1 1 1 1 1 1
1 1 1 1 1 1
NGC 4382 NGC 4459
+"+•
- • - NGC 4526
— • —
i i 1 1 1 1
i i ' • • F • 111 | i i l l
•
•
•
•••••
NGC 4649 NGC 5353
1 10
1 ' — I ' — I I I I 1
1 ' — I
1 10 1 10
r [kpc]
Figure 4.7: Entropy profiles (see equation 4.12) derived from deprojected spectra.
4.2.4 Entropy
Another physical quantity which describes the thermodynamic state of a hot atmosphere
is entropy. Its broadly adopted definition i n this context comes from the equation
describing the adiabatic process i n a monoatomic ideal gas, pV5
^3
= const. Rewriting
this relation in terms of temperature and electron density, the constant is proportional
to K, our measure of entropy, where
K = kBTn-2/3
. (4.12)
This definition relates to the thermodynamic entropy S as AS = 3/21n.KT. Gravitationally
stratified atmosphere i n hydrostatic equilibrium should have the radial entropy
profile monotonically outwardly rising, while flat or decreasing trend signifies convectively
unstable environment. This results from the Schwarzschild stability criterion,
4.2 Thermodynamic properties 30
VK • Vp < 0, where the gas pressure p is radially decreasing due to gravitational
stratification (see the previous section). A detailed derivation of this condition can be
found i n e.g. Pringle and K i n g (2007).
The entropy of all analysed galaxies is separately plotted i n F i g . 4.7 and shows a
visible difference i n the slope of N G C 4649 when compared to our SOs. For a more
robust comparison with elliptical galaxies, the entropy profiles of SOs are plotted
in F i g . 4.8, together w i t h results obtained for the sample of 49 elliptical galaxies
presented b y Lakhchaura et al. (2018) again. The extent of H a emitting gas has
been searched for and observed only i n two galaxies i n our sample, N G C 4459 and
N G C 4526. A direct comparison for the remaining four objects is thus possible only
with the (reasonable) assumption that the colder phases are accompanied by ionized
gas w i t h similar morphology. Another system with observed H a and rotating [C n]
discs, a n d thus having the presumed properties, a lenticular galaxy N G C 7049, is
bridging together with N G C 4459 and N G C 4526 the two samples. Its profile, instead
of following the extended cool gas region of ellipticals, lies above it, similarly to our
SOs. Furthermore, the mean central entropy of the lenticulars lies above that observed
in elliptical galaxies.
t 1 1 1—I—I—I—r-| 1 1 1 1—I—I—I—rJ
1 1 1 1—I—I—I—I p
- Extended Cool Gas
1 10 100
r[kpc]
Figure 4.8: Entropy profiles of SO galaxies in this study (solid lines), N G C 7049, an SO
from previous work (Jurahova et al., 2019, black dotted line) and a sample elliptical galaxies
distinguished by the extent of cool gas of Lakhchaura et al. (2018). For the ellipticals, lines
signify median profiles and surrounding shaded regions the median absolute deviations.
4.2 Thermodynamic properties 31
Purely gravitational heating would result i n a profile given by K oc r 1 1
, which is
usually not observed due to the contribution of AGN-related heating and SNe. These
processes centrally increase the gas entropy, flattening the whole profile i n isolated
galaxies or central regions of galaxy clusters to K oc r a 6 7
(Panagoulia et al., 2014;
Babyk et al., 2018b). To quantify the amount of flattening i n rotating atmospheres,
we fitted the entropy profiles of all SO galaxies in our sample with a power-law model,
together with a fast-rotator N G C 1961 and including also SO galaxies N G C 4477 (Li
et al., 2018) and N G C 7049 (Jurahova et al., 2019), and an E 2 galaxy N G C 6868
(Werner et al., 2014; Lakhchaura et al., 2018), all having rotating discs of warm/cold
gas indicative of rotational support. The resulting profile yields a power-law index
r = 0.46 ± 0.05 and the corresponding best fitting curve is plotted together with all
used data in F i g . 4.9.
- i 1 1 1—i—r l 1 1—i—rr0.46
± 0.05
10
+ this study
-f- NGC 4477
4" NGC 6868
+ NGC 7049
r[kpc]
Figure 4.9: Entropy profiles of rotationally supported galaxies in this study together with
profiles of N G C 4477 (Li et al., 2018), N G C 7049 (Juraiiova et al., 2019), and N G C 6868
(Werner et al., 2014; Lakhchaura et al., 2018). The black line represents a best-fitting power
law model determined from a fit of all plotted data points.
4.3 Thermal stability 32
4.3 T h e r m a l stability
Higher central entropy suggests recent heating of the atmospheres. A G N activity has
been confirmed i n N G C 3665 (see section 2.3) a n d Chandra observations show a n
X-ray point source emission i n the centre of N G C 3607, N G C 3665, N G C 4459, and
N G C 4526. Therefore, to address the thermal stability of the gas, we computed cooling
time profiles using the definition (1.1), which are shown in F i g . A . 4 in the Appendix.
In addition to cooling time alone, profiles of cooling time to free-fall time ratio
were derived from the observed thermodynamic properties. The free-fall time (1.2)
was computed using gravitational acceleration
1 dp
p dr
1 dp
(4.13)
nmn/j, d r '
and thus under the assumption of hydrostatic equilibrium. The resulting values are
presented i n F i g . 4.10 for all eight studied objects and separately in the Appendix,
Fig. A . 3 . The tcoo\/ts « 10 boundary is visualised through the dashed grey line and is
exceeded at all radii, in consistency with observations of other early-type galaxies and
galaxy clusters (Voit et ah, 2018). We note that the tcoo\/ts ratio is fundamentally
independent of radius and thus no common trend is expected nor observable in Fig. 4.10.
Computation of the C-ratio (1.3) requires knowledge of the velocity dispersion
o~v,L at the distance where the turbulence is injected, or the injection scale length, L.
According to Gaspari et al. (2018), this distance can be estimated as a diameter of
100 -
10
-I 1—I—I—I—\
NGC 4382
_j I I I I I I I I I I
10
r[kpc]
Figure 4.10: Ratio of cooling time and free-fall time of all studied galaxies. The threshold of
tcooi/tfi ~ 10 (see section 1.3) is visualised as a dashed grey line.
4.3 Thermal stability 33
the cold/warm phase, and the corresponding UVJL can be obtained by extrapolating
from measured av of the cold gas. For galaxies i n our sample, these measurements
have not been published to date. Nevertheless, to have at least an estimate of
this condensation parameter, we adopt the value of velocity dispersion measured i n
N G C 7049, aVtL = 3610ns-1
, and use it for our calculations. T h e results computed
for SO galaxies with discs of cold gas are plotted in F i g . 4.11 and show that conditions
in these hot haloes are i n favour of condensation from the hot phase.
I
NGC 3607
• NGC 3665
NGC 4459
NGC 4526
NGC 5353
-1 1 1 1 1 1 1 1—|—
5 -
2 -
0.5 -
10
r [kpc]
Figure 4.11: C-ratio (see equation 1.3) of SO galaxies possessing cold gas. Grey region
represents the la confidence region (from hydrodynamical simulations; Gaspari et al., 2018)
signifying conditions for development of multiphase condensation.
C H A P T E R 5
Discussion
5.1 H o t gas m o r p h o l o g y a n d mass
Projected flattening of the hot atmospheres has been found to be similar to or smaller
than that of stellar component in all studied objects, in accord with hydrodynamical
simulations of Negri et al. (2014). It needs to be emphasized that the ellipticity has
been measured on X-ray images (background and point sources subtracted), with
contamination of remaining X-ray emission from unresolved L M X B s , which could
further amplify the observed flattening. However, given that the X-ray emission at
0.3 — 2.0 keV is dominated by the hot gas, the ellipticity is primarily given by the
diffuse gas. Remarkably, the principal axes of the ellipsoidal isophotes fitted to X-ray
and optical emission are aligned within the measured uncertainties. This finding,
together with the non-zero ellipticity, is consistent with ordered rotation of the X-ray
atmospheres in a generally rounder total gravitational potential.
The assumption of spherical symmetry is therefore not formally accurate. However,
the intrinsic scatter given by the quality of data limited by the number of obtained
counts and also the spatial resolution of XMM-Newton reduces the effects of this
simplification. Furthermore, it is important to keep i n m i n d that the ellipticity of
the X-ray atmospheres is low and that the logarithmic scale i n F i g . 4.1 emphasizes
features beyond azimuthal symmetry. Overall, the assumption of spherical symmetry
is not expected to affect our results significantly.
The amount of observed hot gas i n SO galaxies, which has been derived from
deprojected densities out to 2 — 6 Re, varies from 108
to 5 x 1 O 9
M 0 . The largest
amount has been found in N G C 4382 and is comparable to hot gas content in N G C 1961,
while the hot gas mass of the massive slow-rotating elliptical is about two times larger.
Negri et al. (2014) showed that rotationally supported galaxies of the same mass
have, besides lower X-ray luminosity and emission-weighted temperatures, also a lower
amount of hot gas. Here, a comparison based on the same approach is not possible, as
within this small sample the galaxies vary both in rotation and the total galaxy mass
and could have also been affected by the surrounding environment.
5.1.1 Diffuse emission near N G C 4459
A source of diffuse X-ray emission near the hot atmosphere of N G C 4459 is shown in the
left panel of Fig. 5.1. The projected direction of motion of N G C 4459 within the Virgo
34
5.1 Hot gas morphology and mass 35
I I I i 1 1
0 0.9 4.3 18 72 290
Figure 5.1: Left: Background subtracted, exposure corrected and adaptively smoothed X-ray
image centred on diffuse emission near N G C 4459 and extracted from energy range 0.3 —2.0keV,
at which most of the emission can be observed. The colour bar, showing the number of received
counts, allows a qualitative comparison of flux from N G C 4459 and the diffuse source. Right:
geometrically corresponding region of the sky observed in SDSS (data release 9). No nearby
counterpart is observable, but the faint red galaxies could signify a distant cluster.
cluster, as traced by ram-pressure stripped gas (see Fig. 4.1), and the direction from the
galaxy to the diffuse source form an angle of ~ 65°. This extended emission shows an
enhancement in brightness towards its centre (at approximately a = 12h
2 8 m
51s
, 5 =
+13° 59' 36"). The total number of observed net counts from all instruments and both
observations is 8 972 with an energy flux / ~ 2.5 x 1 0 - 9
e r g s - 1
m - 2
. However, no
lines are visible i n the spectrum, which complicates a determination of the redshift.
No counterpart has been observed in U V (Galex), near-infrared (2MASS observations
in J , H, K bands), mid-infrared (Spitzer), far-infrared (Herschel P A C S at 70 urn
and 160 pun), or sub-millimetre wavelengths {Herschel S P I R E at 250 pm, 350 pm, and
500 pm).
In the right panel of Fig. 5.1, an optical image from SDSS covering the same area
as the X-ray image reveals a number of distant red galaxies. To test whether the
X-ray emission could belong to a galaxy cluster with galaxies observable by SDSS, we
calculated luminosity for a maximum redshift at which galaxies have been detected,
z ~ 0.7 (Brescia et ah, 2014). The resulting X-ray luminosity Lx ~ 1 x 1 0 4 4
e r g s _ 1
is
common among galaxy clusters (see e.g. Pratt et ah, 2009).1
1
Shortly before handing in the thesis, we found out that this source has been classified in the
XCLASS catalogue as a galaxy cluster based on follow-up observations in optical and infrared bands
and has photometrically determined redshift of z = 0.5 (Ridl et al., 2017). This object has been
assigned a catalogue number 2260 and is listed in an internal database.
5.2 Thermodynamic properties 36
5.2 T h e r m o d y n a m i c properties
The thermodynamic properties derived from the X-ray spectra are dependent on the
validity of our assumption on the chemical composition of the plasma. In fact, the
projected spectra available for our analysis allowed us to constrain the metallicity
in several cases, and we present the obtained best-fitting values i n F i g . A . 5 . Where
measured, some of the profiles show significant variations as a function of radius.
In principle, these features can indeed be present i n hot atmospheres. A decrease
in abundances of heavy elements towards the outskirts of these atmospheres would
be caused by a presence of gas previously unenriched i n stellar evolution, supplied
from the intergalactic medium. A t the same time, a central decrease of the overall
metallicity could be explained by deposition of heavy elements (excluding noble gases)
into dust grains, as has been recently observationally confirmed by Lakhchaura et al.
(2019).
A reliable metallicity determination is, however, problematic. Analysing Chandra
observations, Babyk et al. (2018a) measured the overall metallicity of a hot atmosphere
in N G C 5353 to be Z = (0.17 ± 0.03) ZQ, three times lower than i n each radial bin
presented in this work. Albeit unusually low (e.g. Mernier et al., 2017), their result is
in accordance with findings of Finoguenov et al. (2007) who studied the hot medium
of the whole H C G 68 group. We remind, however, that this outcome is expectable for
an analysis of multi-temperature spectra fitted with a single-temperature model, as
has been thoroughly discussed by Buote (2000).
The conservative approach adopted here, i.e. leaving the metallicity fixed and
constant for all radial bins, is therefore usually performed in the spectral analysis. We
note that a bias in the overall metallicity would affect the derived physical quantities
as follows. A factor of two difference i n the measured and actual metallicity would
result in 25% bias in the density and pressure, and 17% in the gas entropy. Slopes of
radial profiles would be altered by less than 10 % in a presence of metallicity gradients
(Werner et al., 2012).
A s the only spiral galaxy i n our sample is currently star-forming, unresolved
high-mass X-ray binaries and, to a lesser extent, other stellar sources associated with
young stellar populations are also expected to pollute the observed X-ray emission.
Similarly to L M X B s , the composite spectrum of these stellar sources also forms a
power law, but likely with a different index. However, the results of our analysis are
remarkably similar to those presented by Anderson et al. (2016), who also studied
the hot atmosphere of N G C 1961 but allowed the power-law index to vary during the
fitting procedure to account for all stellar X-ray sources. Therefore, we do not expect
our simplification to have a significant effect on the derived physical quantities.
5.2.1 Temperature
Radial azimuthally averaged profiles reveal that the hot gas temperature is systemically
lower in the SO galaxies than in the massive elliptical N G C 4649. This is an expected
outcome, as the virial temperature of less massive SO galaxies is lower and the ordered
stellar motion i n rotating systems leads to less effective heating (in terms of the
5.2 Thermodynamic properties 37
resulting temperature) of the gas ejected i n stellar mass loss. Outwardly decreasing
temperature in N G C 1961 suggests central heating, which can be provided by an A G N
or a higher S N rate, connected to relatively high star formation of ~ 10 M Q y r _ 1
.
A negative temperature gradient is observable also i n N G C 4382. Provided that
this object does indeed not harbour an A G N (see section 2.4), the source of heating
should also be connected to the stellar population, possibly enhanced by a merger
event. The energy input via type l a SNe can be estimated from their expected rate,
which for a galaxy of this stellar mass ( M * = 4 x 101 1
M 0 ; Gallo et ah, 2010) and S F R
(see Table 2.2) corresponds to approximately 0.02 y r _ 1
using a relation from Sullivan
et ah (2006), or 0.01 y r _ 1
, when derived from the S-band luminosity (Pellegrini, 2012).
W i t h kinetic energy of one explosion, i?sNia ~ 1 0 5 1
erg (e.g. Rosswog and Briiggen,
2007), this rate corresponds to a time-averaged energy injection of 6 x 1 0 4 1
e r g s _ 1
or 3 x 10 e r g s - 1
, respectively. Such heating would be sufficient to compensate for
the energy losses of the hot gas if the efficiency of S N heating was at least ~ 12 % or
~ 27%, respectively. A n additional contribution of the merger event that led to the
creation of shells is expected to be of lesser importance. According to cosmological
numerical simulations analysed by Pop et ah (2018), the initial interaction with the
progenitor occurred 4 — 8 Gyr ago and most of the stars were stripped from it ~ 2 Gyr
ago, while the central cooling time measured in N G C 4382 is t c o o i « 0.5 G y r .
A higher overall temperature of N G C 5353 could be attributed to the fact that
this galaxy is the brightest member of a compact group and to other processes within
it (Sun, 2012).
5.2.2 Pressure
Fig. 4.6 shows that the systemically lower pressure cannot be explained solely by
different total galaxy masses. Even after scaling by r2oo> the order of magnitude gap
between profiles of lenticulars and ellipticals is still present at all radii. In F i g . 5.2,
other general properties are visualised for all studied galaxies: B — V colour index
(Table 2.2), the ratio of stellar rotational velocity and velocity dispersion (Table 2.1)
and X-ray luminosity and gas mass (Table 4.2). B — V colouring reveals that the
pressure is independent of redness of these galaxies, and thus the composition of stellar
populations cannot explain the additional pressure support. Using different colour
indexes, we obtained qualitatively similar results. A t the same time, the absence of
any residual dependence on the rotational velocity (Fig. 4.6) and the ratio of rotational
velocity and velocity dispersion (Fig. 5.2) suggests that this shift cannot be attributed
to rotational support alone.
For E T G s , Pellegrini (2012) claimed that objects with L B < 3 x 101 0
L B , 0 should be
prone to have atmospheres in an outflowing state (see section 1.5). This applies to all
SO galaxies in our sample, perhaps except for a more luminous N G C 4382. The lowered
pressure would then be a direct manifestation of an expanding atmosphere. Similarly,
this would reflect on the trend in X-ray luminosity and hot halo mass, presented in the
bottom panels of Fig. 5.2, but it is necessary to keep in mind that other environmental
effects can play a significant role in affecting the hot gas content. It is also important
to note that we plot the halo mass measured within a radius determined by the data
5.2 Thermodynamic properties 38
quality, which thus does not directly follow the physical properties of the galaxies.
However, using values obtained from integration out to e.g. 5Re of extrapolated
densities, the trend remains visible and is not significantly altered.
The computed values for r2oo,x require some commentary as well. The r2oo,x i n
N G C 4459 is larger than r2oo,GC> while the rotational support or turbulence, which act
against gravity, should result in the opposite outcome. Additionally, other non-thermal
pressure sources, such as magnetic fields or cosmic rays can lead to lower values of
^200,x, if present (see eg. Humphrey et ah, 2013). A combination of additional pressure
support reflects in results of other SO galaxies in our sample. A plausible explanation
for the large r2oo,x in N G C 4459 lies in interactions with the surrounding environment.
The apparent increase in galaxy total mass results from a steeper decrease of pressure
and, at the same time, lower density. F l y i n g through the Virgo cluster, N G C 4459
undergoes ram-pressure stripping and the more loosely bound gas is being removed
from the atmosphere. This interaction can lead to both mentioned effects. Similarly,
the computed r2oo,x of N G C 5353 is likely affected by the intragroup environment.
0.01 0.05 0.01 0.05
r
[r
20o] '"['200]
Figure 5.2: Pressure profiles in scale radius. The line colour is given by B — V colour index
(upper left), the ratio of rotational velocity and velocity dispersion (upper right), hot gas X-ray
luminosity (lower left) and its mass (lower right). The scaled profiles with r2oo,x, determined
from X-ray observations, are shown as dashed lines, while those with known r2oo,GC are drawn
with solid lines. The error bars are omitted for clarity.
5.3 Thermal stability 39
5.2.3 Entropy
Central entropy in most of the studied SO galaxies lies above that of elliptical galaxies,
indicating that a centrally positioned heating mechanism should be present i n these
objects. Low star formation rate suggests that energy injection via winds of young stars
and core-collapse supernovae is most likely incapable of providing sufficient heating. A
plausible source of energy input would be connected to central A G N or type l a SNe,
presumably with an exception of N G C 4382, where the heating should have origin in
processes excluding the central A G N , as discussed above. However, the hot atmosphere
of this galaxy has a remarkably flat entropy profile which suggests that the halo is
centrally overheated and convectively unstable in general.
The combined entropy profiles of rotating galaxies i n our sample and three other
objects with rotating discs of cold gas has been found to be flatter than what has been
observed in cores of galaxy clusters and other elliptical galaxies. The power-law index
measured here, F = 0.46 ± 0 . 0 5 , differs from the result of Babyk et al. (2018b) published
for a sample of 6 galaxies consisting of spirals and lenticulars, F = 0.74 ± 0.06. Their
sample overlaps with our i n two objects, namely N G C 4382 and N G C 5353, finding
quantitatively similar profiles for the two. Their findings are based on data obtained by
Chandra, performing a slightly different analysis using models provided within X S P E C
spectral fitting package, but employing D S D E P R O J method for deprojection as well.
Overall, the flattening i n entropy profiles suggests relatively stronger heating i n the
rotationally supported atmospheres, possibly reflected in our findings concerning the
systemically lower pressure in these systems.
5.3 T h e r m a l stability
Central cooling times do not exceed ~ 0.5 G y r and remain as low as ~ 1 G y r out to
lOkpc, confirming that feedback is necessary to retain these hot atmospheres. The
value of cooling time at 10 kpc for objects with observed cold gas phases is consistent
with findings of Babyk et al. (2018c). This result, along with direct traces of A G N
activity in several objects in our sample, raises the question of thermal stability of the
gas.
The ratio of cooling time to free-fall time does not fall below ten at any radius,
similarly to isolated ellipticals or brightest cluster galaxies (see e.g. Voit et al., 2018).
In addition to the recent findings suggesting that the tC ooiAff ~ 10 represents rather
a limit than a threshold for cooling, there are other reasons for not ruling out the
possibility that the hot gas undergoes cooling via thermal instabilities: The free-fall
time has been calculated under the assumption of hydrostatic equilibrium and, more
importantly, the rotationally supported cooling clumps of gas would not be subjected
to a free fall in the radial direction, owing to angular momentum conservation.
The calculation of C-ratio shows that the turbulence should be capable of generating
density fluctuations prone to cooling. This outcome, yet consistent with the observed
warm and cold phase discs, is dependent on our assumption of present velocity
dispersion as the te ddy scales with <J~\. The velocity dispersion constrained from
resonant scattering and line broadening has been measured by Ogorzalek et al. (2017)
5.3 Thermal stability 40
in several elliptical galaxies using spectra from XMM-Newton Reflection Grating
Spectrometer. The best-fitting 3D av obtained for their whole sample is approximately
1 9 0 k m s _ 1
, but with a large spread around this value for individual objects including
also dispersions consistent w i t h the value used here. This has been constrained for
spectral extraction region width of ~ 5kpc centred at the galaxy core, which is
comparable to our systems, but neither of those studied by Ogorzalek et al. is a
fast rotator. Although these predictions for the development of condensation are not
strictly confined to tC ooiAeddy = 1, the similarity of these two time-scales is required
in order to observe the cooling multiphase gas. The velocity dispersion used i n our
calculations could, in reality, be significantly larger, and so would be the C-ratio. To
obtain the true value of this parameter, spectroscopic measurements of the cold discs
of e.g. the [C n]A157 [im line, are necessary.
The cold gas reservoir could in principle be replenished by stellar ejecta. According
to e.g. Voit and Donahue (2011), the gas ejected from stars should heat up and mix
with the hot phase in systems with as low central density and star formation rate as
in SO galaxies studied here. Contrarily, the presence of cold gas with P A H molecules
suggests a non-negligible role of winds of asymptotic giant branch stars ( A G B ) i n
cold gas deposition. A plausible explanation combining the precipitation from hot
atmospheres and A G B winds has been recently presented by L i et al. (2019). Based on
3D hydrodynamical simulations, they suggest that cooling from the hot phase could
be induced in the mixing layer of the dusty stellar wind and the surrounding hot gas,
leading to preservation of these fragile particles.
C H A P T E R 6
Summary and conclusions
In this thesis, we present a study of X-ray emitting gaseous haloes of six lenticular
galaxies. The sample was selected based on the kinematic state of these objects i n
order to study hot atmospheres subjected to effects of significant angular momentum.
In Chapter 1, we introduce hot atmospheres of galaxies, focusing on their connection
to colder gas phases and expected effects connected to their rotation. In Chapter 2,
the studied sample is presented, with a primary focus on properties related to hot
atmospheres, such as evidence for a presence of the A G N feedback, cold gas deposits,
or star formation. Two more objects are added, a fast-rotating spiral galaxy and an
elliptical in which the rotational support is negligible, both having extended emission
from the hot gas. Chapter 3 is dedicated to a description of the analysis of X-ray
spectra obtained by XMM-Newton E P I C instruments. A s the quality of data allowed
us to investigate the hot gas properties in several radial bins in each galaxy, we perform
a deprojection analysis of these spectra and present the derived radial profiles i n
Chapter 4.
We find an alignment between the hot gas and stellar ellipticities with the X ray
ellipticity generally lower than that of stars, i n an excellent agreement with our
assumption of rotational support in these hot atmospheres and theoretical predictions.
The combined entropy profile of seven rotating galaxies and three galaxies with
indications for rotating atmospheres has been found to be flatter than what is observed
in non-rotating ellipticals, suggesting relatively stronger heating in the central regions
of these systems. We find a systemically lower gas pressure for all rotating galaxies in
our sample when compared to non-rotating ellipticals. Remarkably, this dichotomy
remains pronounced after scaling by total galaxy mass. We emphasize that this feature
is common to all studied objects, having different properties related to manifestations
of A G N activity, star formation rate or interaction with the surrounding environment.
This outcome can be interpreted as a manifestation of outflows present i n these
atmospheres, in agreement with theoretical predictions of Pellegrini (2012).
Investigating the relation between hot haloes and the observed cold gas phases, we
present the derived dimensionless parameters broadly adopted as being related to the
thermal stability of hot atmospheres. The obtained ratio of cooling time and free-fall
time, tcooiAff ^ 10 in all objects, is consistent with observations of ellipticals and even
galaxy clusters. We also estimate the ratio of the cooling and turbulent time-scales
and conclude that from the available data we cannot rule out that the discs of cold
gas present in these objects have condensed out from the hot haloes.
41
A P P E N D I X A
Supplementary material
NGC 1961
NGC 4382
HI
NGC 4649
20 40
NGC 3607
NGC 4459
NGC 5353
0 20 40
r[kpc]
NGC 3665
NGC 4526
20 ~40~
Figure A . l : Deprojected temperature profiles with metallicity fixed at 0.5 Z Q with the
exception of N G C 4649 for which the metallicity is 0.7 Z 0 .
42
A Supplementary material 43
-l .
IQ"3
,
-1 .
IQ"3
,
-1 .
10-2
-.
IQ"3
-.
NGC 1961 NGC 3607
NGC 4382 - NGC 4459
NGC 4649 -
TTTT
10
NGC 5353
10
r[kpc]
NGC 3665
NGC 4526
10
Figure A.2: Particle density profiles derived from deprojected spectra.
A Supplementary material 44
100
10
NGC 1961
•
100 :
10
NGC 4382
100 :
10
NGC 4649
NGC 3607
" T
NGC 4459
++
NGC 5353
10 1 10
r[kpc]
NGC 3665
+
NGC 4526
i-i-T
10
Figure A.3: Ratio of cooling time and free-fall time. The threshold of tcoo\/ts
visualised as a dashed grey line.
10 is
A Supplementary material 45
NGC 1961
10 -
1 -:
0.1
10
NGC 3607
NGC 4382
1 -.
0.1 NGC
4459
NGC 4649
10
0.1
NGC 5353
10 1 10
r [kpc]
NGC 3665
NGC 4526
10
Figure A.4: Cooling time profiles derived from deprojected spectra.
A Supplementary material 46
t
NGC 1961 NGC 3607
+
NGC 3665
NGC 4382 NGC 4459
1+
NGC 4526
NGC 4649
4-
1 —
4
NGC 5353
+4
r [Reff]
Figure A.5: Metallicity profiles derived from projected spectra.
Bibliography
A . B . A l a b i , D . A . Forbes, A . J . Romanowsky, J . P. Brodie, J . Strader, J. Janz,
C. Usher, L . R. Spitler, S. Bellstedt, and A . Ferre-Mateu. The S L U G G S survey:
dark matter fractions at large radii and assembly epochs of early-type galaxies from
globular cluster kinematics. M N R A S , 468:3949-3964, J u l 2017.
M . E . Anderson, E . Churazov, and J . N . Bregman. A deep X M M - N e w t o n study of
the hot gaseous halo around N G C 1961. M N R A S , 455(l):227-243, Jan 2016.
Iu. V . Babyk, B . R. M c N a m a r a , P. E . J . Nulsen, M . T. Hogan, A . N . Vantyghem,
H. R. Russell, F . A . Pulido, and A . C. Edge. X - R a y Scaling Relations of Early-type
Galaxies. A p J , 857(1):32, A p r 2018a.
Iu. V . Babyk, B . R. McNamara, P. E . J. Nulsen, H . R. Russell, A . N . Vantyghem, M . T.
Hogan, and F . A . Pulido. A Universal Entropy Profile for the Hot Atmospheres of
Galaxies and Clusters within R 2500- A p J , 862(1):39, Jul 2018b.
Iu. V . Babyk, B . R . McNamara, P. D . Tamhane, P. E . J . Nulsen, H . R . Russell, and
A . C . Edge. Origins of molecular clouds in early-type galaxies. arXiv e-prints, art.
arXiv:1810.11465, Oct 2018c.
D. J. Barnes, R. Kannan, M . Vogelsberger, and F . Marinacci. Radiative A G N feedback
on a moving mesh: the impact of the galactic disc and dust physics on outflow
properties. arXiv e-prints, art. arXiv:1812.01611, Dec 2018.
H . Boehringer, W . Voges, A . C. Fabian, A . C . Edge, and D . M . Neumann. A R O S A T
H R I study of the interaction of the X-ray emitting gas and radio lobers of N G C
1275. M N R A S , 264:L25-L28, Oct 1993.
A . Bogdan, W . R. Forman, M . Vogelsberger, H . Bourdin, D . Sijacki, P. Mazzotta, R. P.
Kraft, C . Jones, M . Gilfanov, E . Churazov, and L . P. David. Hot X - R a y Coronae
around Massive Spiral Galaxies: A Unique Probe of Structure Formation Models.
A p J , 772:97, A u g 2013.
M . Brescia, S. Cavuoti, G . Longo, and V . De Stefano. A catalogue of photometric
redshifts for the S D S S - D R 9 galaxies. A & A , 568:A126, August 2014.
T. J. Bridges and J. A . Irwin. Molecular gas in the Perseus cooling flow galaxy, N G C
1275. M N R A S , 300(4):967-976, Nov 1998.
47
Bibliography 48
M . J. I. Brown, B . T. Jannuzi, D . J. E . Floyd, and J. R. Mould. The Ubiquitous Radio
Continuum Emission from the Most Massive Early-type Galaxies. A p J , 731(2):L41,
Apr 2011.
D. A . Buote. X-ray evidence for multiphase hot gas with nearly solar Fe abundances
in the brightest groups of galaxies. M N R A S , 311(l):176-200, Jan 2000.
A . Capetti, P. Kharb, D . J . Axon, D . Merritt, and R. D . Baldi. A Very Large Array
Radio Survey of Early-Type Galaxies in the Virgo Cluster. A J , 138:1990-1997,
December 2009.
M . Cappellari, E . Emsellem, D . Kraj novic, R. M . McDermid, N . Scott, G . A . Verdoes
Kleijn, L . M . Young, K . Alatalo, R. Bacon, L . Blitz, M . Bois, F . Bournaud, M . Bureau,
R. L . Davies, T. A . Davis, P. T. de Zeeuw, P . - A . Due, S. Khochfar, H . Kuntschner,
P . - Y . Lablanche, R. Morganti, T. Naab, T. Oosterloo, M . Sarzi, P. Serra, and
A . - M . Weijmans. The A T L A S 3 0
project - I. A volume-limited sample of 260 nearby
early-type galaxies: science goals and selection criteria. M N R A S , 413:813-836, May
2011.
R. Carrillo, J . Masegosa, D . Dultzin-Hacyan, and R. Ordonez. A Multifrequency
Catalog of L I N E R s . Rev. Mexičana Astron. Astrofis., 35:187, Oct 1999.
J. A . Carter and A . M . Read. The X M M - N e w t o n E P I C background and the production
of background blank sky event files. A & A , 464:1155-1166, March 2007.
W . Cash. Parameter estimation in astronomy through application of the likelihood
ratio. A p J , 228:939-947, March 1979.
A . Cavaliere and R. Fusco-Femiano. X-rays from hot plasma in clusters of galaxies.
A & A , 49:137-144, M a y 1976.
A . Cavaliere and R. Fusco-Femiano. The Distribution of Hot Gas in Clusters of
Galaxies. A & A , 70:677, November 1978.
E . Churazov, M . Brüggen, C. R. Kaiser, H . Böhringer, and W . Forman. Evolution of
Buoyant Bubbles in M 8 7 . A p J , 554(l):261-273, Jun 2001.
E . Churazov, S. Sazonov, R. Sunyaev, W . Forman, C . Jones, and H . Böhringer.
Supermassive black holes in elliptical galaxies: switching from very bright to very
dim. M N R A S , 363(1):L91-L95, Oct 2005.
S. Cielo, R. Bieri, M . Volonteri, A . Y . Wagner, and Y . Dubois. A G N feedback compared:
jets versus radiation. M N R A S , 477(1):1336-1355, Jun 2018.
F . Combes, L . M . Young, and M . Bureau. Molecular gas and star formation in the
S A U R O N early-type galaxies. M N R A S , 377(4) :1795-1807, Jun 2007.
F . Combes, A . J . Baker, E . Schinnerer, S. Garcia-Burillo, L . K . Hunt, F . Boone,
A . Eckart, R. Neri, and L . J. Tacconi. Molecular gas in NUclei of GAlaxies ( N U G A ) .
XII. The head-on collision in N G C 1961. A & A , 503:73-86, August 2009.
Bibliography 49
J. J . Condon, W . D . Cotton, and J . J. Broderick. Radio Sources and Star Formation
in the Local Universe. A J , 124(2):675-689, A u g 2002.
C. J . Conselice, III Gallagher, J. S., and R . F . G . Wyse. O n the Nature of the N G C
1275 System. A J , 122(5), Nov 2001.
T. A . Davis, M . Bureau, M . Cappellari, M . Sarzi, and L . Blitz. A black-hole mass
measurement from molecular gas kinematics in NGC4526. Nature, 494:328-330, Feb
2013.
T. A . Davis, J . E . Greene, C.-P. M a , J . P. Blakeslee, J . M . Dawson, V . Pandya,
M . Veale, and N . Zabel. The M A S S I V E survey - X L What drives the molecular gas
properties of early-type galaxies. M N R A S , page 847, M a r 2019.
R. L . Dickman, Ronald L . Snell, and F . Peter Schloerb. Carbon Monoxide as an
Extragalactic Mass Tracer. A p J , 309:326, Oct 1986.
M . Donahue, G . E . de Messieres, R. W . O'Connell, G . M . Voit, A . Hoffer, B . R.
McNamara, and P. E . J . Nulsen. Polycyclic Aromatic Hydrocarbons, Ionized Gas,
and Molecular Hydrogen in Brightest Cluster Galaxies of Cool-core Clusters of
Galaxies. A p J , 732(1):40, M a y 2011.
B . T . Draine. Physics of the Interstellar and Intergalactic Medium. 2011.
R. J . H . D u n n , S. W . Allen, G . B . Taylor, K . F . Shurkin, G . Gentile, A . C . Fabian,
and C . S. Reynolds. The radio properties of a complete, X-ray selected sample of
nearby, massive elliptical galaxies. M N R A S , 404:180-197, M a y 2010.
E . Emsellem, M . Cappellari, D . Krajnovic, K . Alatalo, L . Blitz, M . Bois, F . Bournaud,
M . Bureau, R. L . Davies, T. A . Davis, P. T. de Zeeuw, S. Khochfar, H . Kuntschner,
P.-Y. Lablanche, R . M . McDermid, R . Morganti, T. Naab, T. Oosterloo, M . Sarzi,
N . Scott, P. Serra, G . van de Ven, A . - M . Weijmans, and L . M . Young. The A T L A S 3 D
project - III. A census of the stellar angular momentum within the effective radius
of early-type galaxies: unveiling the distribution of fast and slow rotators. M N R A S ,
414:888-912, June 2011.
E S A : X M M - N e w t o n S O C . X M M - N e w t o n Users Handbook, 2018. U R L https:
//xmm-tools.cosmos.esa.int/external/xmm_user_support/documentation/
uhb/. [Online; accessed 2019-05-14].
P. B . Eskridge, G . Fabbiano, and D . - W . K i m . A Multiparametric Analysis of the
Einstein Sample of Early-Type Galaxies. I. Luminosity and I S M Parameters. The
Astrophysical Journal Supplement Series, 97:141, M a r 1995.
A . C . Fabian. Cooling Flows i n Clusters of Galaxies. Annual Review of Astronomy
and Astrophysics, 32:277-318, Jan 1994.
A . C . Fabian and P. E . J . Nulsen. Subsonic accretion of cooling gas i n clusters of
galaxies. M N R A S , 180:479-484, A u g 1977.
Bibliography 50
G . J . Ferland, A . C . Fabian, and R . M . Johnstone. The Physical Conditions W i t h i n
Dense Cold Clouds in Cooling Flows. M N R A S , 266:399, January 1994.
L. Ferrarese and D . Merritt. A Fundamental Relation between Supermassive Black
Holes and Their Host Galaxies. A p J , 539(1):L9-L12, A u g 2000.
L. Ferrarese, P. Cote, A . Jordan, E . W . Peng, J. P. Blakeslee, S. Piatek, S. M e i ,
D. Merritt, M . Milosavljevic, J . L . Tonry, and M . J. West. The A C S Virgo Cluster
Survey. V I . Isophotal Analysis and the Structure of Early-Type Galaxies. The
Astrophysical Journal Supplement Series, 164:334-434, Jun 2006.
G . B . Field. Thermal Instability. A p J , 142:531, August 1965.
A . Finoguenov and C . Jones. Chandra Observation of M84, a Radio Lobe Elliptical
Galaxy in the Virgo Cluster. A p J , 547(2):L107-L110, Feb 2001.
A . Finoguenov, T. J. Ponman, J. P. F . Osmond, and M . Zimer. X M M - N e w t o n study of
0.012 &lt; z &lt; 0.024 groups - I. Overview of the I G M thermodynamics. M N R A S ,
374:737-760, Jan 2007.
W . Forman, P. Nulsen, S. Heinz, F . Owen, J . Eilek, A . Vikhlinin, M . Markevitch,
R. Kraft, E . Churazov, and C . Jones. Reflections of Active Galactic Nucleus
Outbursts in the Gaseous Atmosphere of M87. A p J , 635(2):894-906, Dec 2005.
A . Fruscione, J. C. McDowell, G . E . Allen, N . S. Brickhouse, D . J. Burke, J. E . Davis,
N . Durham, M . Elvis, E . C . Galle, D . E . Harris, D . P. Huenemoerder, J. C . Houck,
B . Ishibashi, M . Karovska, F . Nicastro, M . S. Noble, M . A . Nowak, F . A . Primini,
A . Siemiginowska, R. K . Smith, and M . Wise. C I A O : Chandra's data analysis
system. In Society of Photo-Optical Instrumentation Engineers (SPIE) Conference
Series, volume 6270 of Proc. SPIE, page 62701V, June 2006.
E . Gallo, T. Treu, P. J. Marshall, J . - H . Woo, C. Leipski, and R. Antonucci. A M U S E Virgo.
II. Down-sizing in Black Hole Accretion. A p J , 714(l):25-36, M a y 2010.
M . Gaspari and E . Churazov. Constraining turbulence and conduction in the hot I C M
through density perturbations. A & A , 559:A78, Nov 2013.
M . Gaspari, P. Temi, and F . Brighenti. Raining on black holes and massive galaxies:
the top-down multiphase condensation model. M N R A S , 466(1) :677-704, A p r 2017.
M . Gaspari, M . McDonald, S. L . Hamer, F . Brighenti, P. Temi, M . Gendron-Marsolais,
J. Hlavacek-Larrondo, A . C . Edge, N . Werner, P. Tozzi, M . Sun, J. M . Stone, G . R.
Tremblay, M . T. Hogan, D . Eckert, S. Ettori, H . Y u , V . Biffi, and S. Planelles.
Shaken Snow Globes: Kinematic Tracers of the Multiphase Condensation Cascade
in Massive Galaxies, Groups, and Clusters. A p J , 854:167, February 2018.
G. Gavazzi, G . Consolandi, S. Pedraglio, M . Fossati, M . Fumagalli, and A . Boselli. H a
imaging observations of early-type galaxies from the A T L A S 3 0
survey. A & A , 611:
A28, M a r 2018.
Bibliography 51
K . Gebhardt, R. Bender, G . Bower, A . Dressier, S. M . Faber, A . V . Filippenko,
R. Green, C Grillmair, L . C Ho, J. Kormendy, T. R. Lauer, J. Magorrian, J. Pinkney,
D. Richstone, and S. Tremaine. A Relationship between Nuclear Black Hole Mass
and Galaxy Velocity Dispersion. A p J , 539(1):L13-L16, A u g 2000.
G. Giuricin, C . Marinoni, L . Ceriani, and A . Pisani. Nearby Optical Galaxies: Selection
of the Sample and Identification of Groups. A p J , 543(1):178-194, Nov 2000.
C. F . Goullaud, J. B . Jensen, J. P. Blakeslee, C.-P. M a , J. E . Greene, and J. Thomas.
The M A S S I V E Survey. I X . Photometric Analysis of 35 High-mass Early-type Galaxies
with H S T W F C 3 / I R . A p J , 856:11, M a r 2018.
R. Grossova, N . Werner, K . Rajpurohit, F . Mernier, K . Lakhchaura, K . Gabanyi,
R. E . A . Canning, P. Nulsen, F . Massaro, M . Sun, T . Connor, A . K i n g , S. W .
Allen, R . Lyn-Salmon Frisbie, M . Donahue, and A . C . Fabian. Powerful A G N jets
and unbalanced cooling in the hot atmosphere of IC 4296. arXiv e-prints, art.
arXiv:1903.03198, M a r 2019.
K . Gúltekin, D . O. Richstone, K . Gebhardt, S. M . Faber, T . R. Lauer, R. Bender,
J. Kormendy, and J . Pinkney. Is There a Black Hole in N G C 4382? A p J , 741:38,
Nov 2011.
S. Haan, E . Schinnerer, C . G . Mundell, S. Garcia-Burillo, and F . Combes. Atomic
Hydrogen Properties of Active Galactic Nuclei Host Galaxies: H I in 16 Nuclei of
Galaxies ( N U G A ) Sources. A J , 135(1):232-257, Jan 2008.
A . I. Harris, A . J. Baker, S. G . Zonak, C. E . Sharon, R. Genzel, K . Rauch, G . Watts,
and R. Creager. C O J = 1-0 Spectroscopy of Four Submillimeter Galaxies with the
Zpectrometer on the Green Bank Telescope. A p J , 723(2):1139-1149, Nov 2010.
C. M . Harrison. Impact of supermassive black hole growth on star formation. Nature
Astronomy, 1:0165, J u l 2017.
P. Hickson. Systematic properties of compact groups of galaxies. A p J , 255:382-391,
A p r 1982.
M . T . Hogan, B . R. M c N a m a r a , F . Pulido, P. E . J . Nulsen, H . R. Russell, A . N .
Vantyghem, A . C . Edge, and R . A . M a i n . Mass Distribution in Galaxy Cluster
Cores. A p J , 837(1):51, M a r 2017.
P. J . Humphrey, D . A . Buote, F . Brighenti, K . Gebhardt, and W . G . Mathews.
Reconciling stellar dynamical and hydrostatic X-ray mass measurements of an
elliptical galaxy with gas rotation, turbulence and magnetic fields. M N R A S , 430(3):
1516-1528, A p r 2013.
J. A . Irwin, A . E . Athey, and J. N . Bregman. X - R a y Spectral Properties of Low-Mass
X - R a y Binaries in Nearby Galaxies. A p J , 587:356-366, A p r i l 2003.
Bibliography 52
N . James, D . - W . K i m , G . Fabbiano, D . Forbes, and A . Alabi. The Mass of the Globular
Cluster Systems of Early Type Galaxies as Proxy for the Total Galaxy Mass. arXiv
e-prints, art. arXiv:1810.09475, Oct 2018.
F. Jansen, D . Lumb, B . Altieri, J. Clavel, M . Ehle, C . E r d , C . Gabriel, M . Guainazzi,
P. Gondoin, R. Much, R. Munoz, M . Santos, N . Schartel, D . Texier, and G . Vacanti.
X M M - N e w t o n observatory. I. The spacecraft and operations. A & A , 365:L1-L6,
January 2001.
T. H . Jarrett, T. Chester, R . Cutri, S. E . Schneider, and J . P. Huchra. The 2 M A S S
Large Galaxy Atlas. A J , 125(2):525-554, Feb 2003.
A . Jurahova, N . Werner, M . Gaspari, K . Lakhchaura, P. E . J. Nulsen, M . Sun, R. E . A .
Canning, S. W . Allen, A . Simionescu, J. B . R. Oonk, T. Connor, and M . Donahue.
Cooling in the X-ray halo of the rotating, massive early-type galaxy N G C 7049.
M N R A S , 484(2):2886-2895, A p r 2019.
J. S. Kaastra, R . Mewe, and H . Nieuwenhuijzen. S P E X : a new code for spectral
analysis of X & U V spectra. In K . Yamashita and T . Watanabe, editors, UV and
X-ray Spectroscopy of Astrophysical and Laboratory Plasmas, pages 411-414, 1996.
P. M . W . Kalberla, W . B . Burton, D . Hartmann, E . M . Arnal, E . Bajaja, R. Morras, and
W . G . L . Poppel. The Leiden/Argentine/Bonn ( L A B ) Survey of Galactic HI. Final
data release of the combined L D S and I A R surveys with improved stray-radiation
corrections. A & A , 440:775-782, September 2005.
T. Kokusho, H . Kaneda, M . Bureau, T. Suzuki, K . Murata, A . Kondo, and M . Yamagishi.
A star formation study of the A T L A S 3 0
early-type galaxies with the A K A R I
all-sky survey. A & A , 605:A74, Sep 2017.
J. Kormendy. A critical review of stellar-dynamical evidence for black holes in galaxy
nuclei. In J. Beckman, L . Colina, and H . Netzer, editors, The Nearest Active Galaxies,
pages 197-218, January 1993.
D. Krajnovic, E . Emsellem, M . Cappellari, K . Alatalo, L . Blitz, M . Bois, F . Bournaud,
M . Bureau, R. L . Davies, T. A . Davis, P. T. de Zeeuw, S. Khochfar, H . Kuntschner,
P.-Y. Lablanche, R . M . McDermid, R . Morganti, T. Naab, T. Oosterloo, M . Sarzi,
N . Scott, P. Serra, A . - M . Weijmans, and L . M . Young. The A T L A S 3 D
project
- II. Morphologies, kinemetric features and alignment between photometric and
kinematic axes of early-type galaxies. M N R A S , 414:2923-2949, July 2011.
M . Krips, A . Eckart, T. P. Krichbaum, J. U . Pott, S. Leon, R. Neri, S. Garcia-Burillo,
F. Combes, F . Boone, A . J . Baker, L . J . Tacconi, E . Schinnerer, and L . K . Hunt.
NUclei of GAlaxies. V . Radio emission in 7 N U G A sources. A & A , 464:553-563, M a r
2007.
C. del P. Lagos, T. A . Davis, C . G . Lacey, M . A . Zwaan, C. M . Baugh, V . GonzalezPerez,
and N . D . Padilla. The origin of the atomic and molecular gas contents of
Bibliography 53
early-type galaxies - I. A new test of galaxy formation physics. M N R A S , 443(2):
1002-1021, Sep 2014.
K . Lakhchaura, N . Werner, M . Sun, R. E . A . Canning, M . Gaspari, S. W . Allen,
T. Connor, M . Donahue, and C . Sarazin. Thermodynamic properties, multiphase
gas, and A G N feedback i n a large sample of giant ellipticals. M N R A S , 481(4):
4472-4504, Dec 2018.
K . Lakhchaura, F . Mernier, and N . Werner. Possible depletion of metals into dust
grains in the core of the Centaurus cluster of galaxies. A & A , 623:A17, M a r 2019.
Y . L i , G . L . Bryan, M . Ruszkowski, G . M . Voit, B . W . O'Shea, and M Donahue. Cooling,
A G N Feedback, and Star Formation in Simulated Cool-core Galaxy Clusters. A p J ,
811(2):73, Oct 2015.
Y . L i , Y . Su, and C . Jones. X-ray cavities i n the hot corona of the lenticular galaxy
N G C 4477. M N R A S , 480(4):4279-4286, Nov 2018.
Y . L i , G . L . Bryan, and E . Quataert. The Fate of A G B W i n d in Massive Galaxies and
the I C M . arXiv e-prints, art. arXiv:1901.10481, Jan 2019.
E . Liuzzo, G . Giovannini, M . Giroletti, and G . B . Taylor. The Bologna complete
sample of nearby radio sources. II. Phase referenced observations of faint nuclear
sources. A & A , 505:509-520, Oct 2009.
K . Lodders, H . Palme, and H . - P . Gail. Abundances of the Elements i n the Solar
System. Landolt Bornstein, 2009.
D. M . Lucero and L . M . Young. A High-resolution Study of the Atomic Hydrogen in
CO-rich Early-type Galaxies. A J , 145:56, M a r 2013.
D . Makarov and I. Karachentsev. Galaxy groups and clouds in the local (z~0.01)
Universe. M N R A S , 412(4):2498-2520, A p r 2011.
D. Makarov, P. Prugniel, N . Terekhova, H . Courtois, and I. Vaughn. H y p e r L E D A . III.
The catalogue of extragalactic distances. A & A , 570:A13, October 2014.
G . A . M a m o n . The nature of the nearest compact group of galaxies from precise
distance measurements. A & A , 486(1):113-117, Jul 2008.
M . M c C o u r t , I. J . Parrish, P. Sharma, and E . Quataert. C a n conduction induce
convection? O n the non-linear saturation of buoyancy instabilities in dilute plasmas.
M N R A S , 413(2):1295-1310, M a y 2011.
M . McDonald, S. Veilleux, D . S. N . Rupke, and R . Mushotzky. O n the Origin of the
Extended H a Filaments in Cooling Flow Clusters. A p J , 721(2) :1262-1283, Oct 2010.
M . McDonald, S. Veilleux, and R. Mushotzky. The Effect of Environment on the
Formation of H a Filaments and Cool Cores i n Galaxy Groups and Clusters. A p J ,
731(1):33, A p r 2011.
Bibliography 54
B . R. McNamara, H . R. Russell, P. E . J. Nulsen, M . T. Hogan, A . C. Fabian, F . Pulido,
and A . C . Edge. A Mechanism for Stimulating A G N Feedback by Lifting Gas in
Massive Galaxies. A p J , 830(2):79, Oct 2016.
F . Mernier, J. de Plaa, J. S. Kaastra, Y . Y . Zhang, H . Akamatsu, L . G u , P. Kosec,
J. M a o , C . Pinto, T. H . Reiprich, J . S. Sanders, A . Simionescu, and N . Werner.
Radial metal abundance profiles in the intra-cluster medium of cool-core galaxy
clusters, groups, and ellipticals. A & A , 603A80, Jul 2017.
T. Monfredini, H . M . Quitian-Lara, F . Fantuzzi, W . Wolff, E . Mendoza, A . F . Lago,
D. A . Sales, M . G . Pastoriza, and H . M . Boechat-Roberty. Destruction and multiple
ionization of P A H s by X-rays in circumnuclear regions of A G N s . arXiv e-prints, art.
arXiv:1808.07626, A u g 2018.
R. Morganti. The many routes to A G N feedback. Frontiers in Astronomy and Space
Sciences, 4:42, Nov 2017.
J. S. Mulchaey, D . S. Davis, R . F . Mushotzky, and D . Burstein. A n X - R a y Atlas of
Groups of Galaxies. The Astrophysical Journal Supplement Series, 145(l):39-64,
Mar 2003.
A . Negri, S. Posacki, S. Pellegrini, and L . Ciotti. The effects of galaxy shape and
rotation on the X-ray haloes of early-type galaxies - II. Numerical simulations.
M N R A S , 445(2):1351-1369, Dec 2014.
N . Neininger, M . Guelin, H . Ungerechts, R. Lucas, and R. Wielebinski. Carbon
monoxide emission as a precise tracer of molecular gas in the Andromeda galaxy.
Nature, 395:871-873, October 1998.
C. P. O'Dea, S. A . Baum, G . Privon, J. Noel-Storr, A . C. Quillen, N . Zufelt, J. Park,
A . Edge, H . Russell, A . C. Fabian, M . Donahue, C. L . Sarazin, B . McNamara, J. N .
Bregman, and E . Egami. A n Infrared Survey of Brightest Cluster Galaxies. II. W h y
are Some Brightest Cluster Galaxies Forming Stars? A p J , 681(2):1035-1045, J u l
2008.
A . Ogorzalek, I. Zhuravleva, S. W . Allen, C . Pinto, N . Werner, A . B . Mantz, R. E . A .
Canning, A . C . Fabian, J . S. Kaastra, and J . de Plaa. Improved measurements
of turbulence in the hot gaseous atmospheres of nearby giant elliptical galaxies.
M N R A S , 472(2):1659-1676, Dec 2017.
K . Onishi, S. Iguchi, T. A . Davis, M . Bureau, M . Cappellari, M . Sarzi, and L . Blitz.
W I S D O M project - I. Black hole mass measurement using molecular gas kinematics
in N G C 3665. M N R A S , 468:4663-4674, J u l 2017.
E . O'Sullivan, F . Combes, P. Salome, L . P. David, A . Babul, J . M . Vrtilek, J . L i m ,
V . Olivares, S. Raychaudhury, and G . Schellenberger. Cold gas in a complete sample
of group-dominant early-type galaxies. A & A , 618A126, Oct 2018.
Bibliography 55
E . K . Panagoulia, A . C . Fabian, and J. S. Sanders. A volume-limited sample of X-ray
galaxy groups and clusters - I. Radial entropy and cooling time profiles. M N R A S ,
438(3)=2341-2354, M a r 2014.
P. Parma, H . R. de Ruiter, C. Fanti, and R. Fanti. V L A observations of low luminosity
radio galaxies. I. Sources with angular size smaller than two arcminutes. Astronomy
and Astrophysics Supplement Series, 64:135-171, A p r 1986.
S. Pellegrini. Hot Gas Flows on Global and Nuclear Galactic Scales. In Dong-Woo
K i m and Silvia Pellegrini, editors, Astrophysics and Space Science Library, volume
378 of Astrophysics and Space Science Library, page 21, Jan 2012.
J. R. Peterson and A . C . Fabian. X-ray spectroscopy of cooling clusters. Phys. Rep.,
427:1-39, A p r i l 2006.
J. R . Peterson, S. M . K a h n , F . B . S. Paerels, J . S. Kaastra, T . Tamura, J . A . M .
Bleeker, C . Ferrigno, and J . G . Jernigan. High-Resolution X - R a y Spectroscopic
Constraints on Cooling-Flow Models for Clusters of Galaxies. A p J , 590(1):207-224,
Jun 2003.
F . Pizzolato and N . Soker. O n the Nature of Feedback Heating i n Cooling Flow
Clusters. A p J , 632(2):821-830, Oct 2005.
A . -R. Pop, A . Pillepich, N . C . Amorisco, and L . Hernquist. Formation and incidence
of shell galaxies in the Illustris simulation. M N R A S , 480(2):1715-1739, Oct 2018.
D . Prasad, P. Sharma, and A . Babul. Cool Core Cycles: Cold Gas and A G N Jet
Feedback in Cluster Cores. A p J , 811(2):108, Oct 2015.
G . W . Pratt, J. H . Croston, M . A r n a u d , and H . Bohringer. Galaxy cluster X-ray
luminosity scaling relations from a representative local sample ( R E X C E S S ) . A & A ,
498(2) :361-378, M a y 2009.
J. E . Pringle and A . King. Astrophysical Flows. 2007.
F . A . Pulido, B . R. M c N a m a r a , A . C . Edge, M . T. Hogan, A . N . Vantyghem, H . R.
Russell, P. E . J. Nulsen, I. Babyk, and P. Salome. The Origin of Molecular Clouds
in Central Galaxies. A p J , 853(2):177, Feb 2018.
M . J. Rees and J . P. Ostriker. Cooling, dynamics and fragmentation of massive gas
clouds: clues to the masses and radii of galaxies and clusters. M N R A S , 179:541-559,
Jun 1977.
B . L . Refsdal, S. M . Doe, D . T. Nguyen, A . L . Siemiginowska, N . R. Bonaventura,
D. Burke, I. N . Evans, J . D . Evans, A . Fruscione, E . C . Galle, J . C . Houck,
M . Karovska, N . P. Lee, and M . A . Nowak. Sherpa: l d / 2 d modeling and fitting
in python. In Gael Varoquaux, Stefan van der Walt, and Jarrod Millman, editors,
Proceedings of the 8th Python in Science Conference, pages 51 - 57, Pasadena, C A
U S A , 2009.
Bibliography 56
J. Ridl, N . Clerc, T. Sadibekova, L . Faccioli, F . Pacaud, J . Greiner, T. Krühler, A . Rau,
M . Salvato, M . L . Menzel, H . Steinle, P. Wiseman, K . Nandra, and J . Sanders.
Cosmology with X M M galaxy clusters: the X - C L A S S / G R O N D catalogue and
photometric redshifts. M N R A S , 468(l):662-684, Jun 2017.
S. Rosswog and M . Brüggen. Introducton to High-Energy Astrophysics. Cambridge
University Press, 2007. I S B N 978-0-521-85769-7.
V . C . Rubin, W . K . Jr. Ford, and M . S. Roberts. Extended rotation curves of highluminosity
spiral galaxies. V . N G C 1961, the most massive spiral known. A p J , 230:
35-39, M a y 1979.
H . R . Russell, J . S. Sanders, and A . C . Fabian. Direct X-ray spectral deprojection of
galaxy clusters. M N R A S , 390:1207-1216, Nov 2008.
P. Saikia, E . Körding, D . L . Coppejans, H . Falcke, D . Williams, R. D . Baldi, I. Mchardy,
and R . Beswick. 15-GHz radio emission from nearby low-luminosity active galactic
nuclei. A & A , 616:A152, Sep 2018.
J. R. Sanchez-Sutil, A . J . Munoz-Arjonilla, J . Marti, J . L . Garrido, D . Perez-Ramirez,
and P. Luque-Escamilla. A catalogue of ultra-luminous X-ray source coincidences
with F I R S T radio sources. A & A , 452:739-742, Jun 2006.
E . Scannapieco and M . Brüggen. Subgrid Modeling of A G N - d r i v e n Turbulence i n
Galaxy Clusters. A p J , 686(2) :927-947, Oct 2008.
P. Sharma, M . McCourt, E . Quataert, and I. J . Parrish. Thermal instability and the
feedback regulation of hot haloes i n clusters, groups and galaxies. M N R A S , 420:
3174-3194, March 2012.
G. S. Shostak, E . Hummel, P. A . Shaver, J . M . van der Hülst, and P. C . van der Kruit.
N G C 1961: Stripping of a supermassive spiral galaxy. A & A , 115:293-307, Nov 1982.
K . Shurkin, R. J . H . Dunn, G . Gentile, G . B . Taylor, and S. W . Allen. Active galactic
nuclei-induced cavities i n N G C 1399 and N G C 4649. M N R A S , 383:923-930, Jan
2008.
A . Simionescu, N . Werner, A . Finoguenov, H . Böhringer, and M . Brüggen. Metal-rich
multi-phase gas i n M 87. A G N - d r i v e n metal transport, magnetic-field supported
multi-temperature gas, and constraints on non-thermal emission observed with
X M M - N e w t o n . A & A , 482(1):97-112, A p r 2008.
S. L . Snowden and K . D . Kuntz. Analysis of X M M - N e w t o n D a t a from Extended
Sources and the Diffuse X-ray Background. In American Astronomical Society
Meeting Abstracts #217, volume 43 of Bulletin of the American Astronomical
Society, page 344.17, January 2011.
Bibliography 57
S. Stierwalt, L . Armus, V . Charmandaris, T . Diaz-Santos, J . Marshall, A . S. Evans,
S. Haan, J . Howell, K . Iwasawa, D . C . K i m , E . J . Murphy, J . A . Rich, H . W . W .
Spoon, H . Inami, A . O. Petric, and V . U . Mid-infrared Properties of Luminous
Infrared Galaxies. II. Probing the Dust and Gas Physics of the G O A L S Sample.
A p J , 790(2):124, A u g 2014.
L. Strüder, U . Briel, K . Dennerl, R. Hartmann, E . Kendziorra, N . Meidinger, E . Pfeffermann,
C . Reppin, B . Aschenbach, W . Bornemann, H . Bräuninger, W . Burkert,
M . Elender, M . Freyberg, F . Haberl, G . Hartner, F . Heuschmann, H . Hippmann,
E. Kastelic, S. Kemmer, G . Kettenring, W . Kink, N . Krause, S. Müller, A . Oppitz,
W . Pietsch, M . Popp, P. Predehl, A . Read, K . H . Stephan, D . Stötter, J. Trimmer,
P. Holl, J . Kemmer, H . Soltau, R. Stötter, U . Weber, U . Weichert, C . von Zanthier,
D . Carathanassis, G . Lutz, R. H . Richter, P. Sole, H . Böttcher, M . Küster,
R. Staubert, A . Abbey, A . Holland, M . Turner, M . Balasini, G . F . Bignami, N . L a
Palombara, G . Villa, W . Buttler, F . Gianini, R. Laine, D . L u m b , and P. Dhez.
The European Photon Imaging Camera on X M M - N e w t o n : The p n - C C D camera.
Astrophys. J., 365T18-L26, January 2001.
M . Sullivan, D . Le Borgne, C. J. Pritchet, A . Hodsman, J. D . Neill, D . A . Howell, R. G .
Carlberg, P. Astier, E . Aubourg, D . Balam, S. Basa, A . Conley, S. Fabbro, D . Fouchez,
J. Guy, I. Hook, R. Pain, N . Palanque-Delabrouille, K . Perrett, N . Regnault, J. Rich,
R. Taillet, S. Baumont, J. Bronder, R. S. Ellis, M . Filiol, V . Lusset, S. Perlmutter,
P. Ripoche, and C. Tao. Rates and Properties of Type l a Supernovae as a Function
of Mass and Star Formation in Their Host Galaxies. A p J , 648(2):868-883, Sep 2006.
M . Sun. Hot gas in galaxy groups: recent observations. New Journal of Physics, 14
(4):045004, A p r 2012.
P. Temi, W . G . Mathews, F . Brighenti, and J . D . Bregman. Dust in Hot Gas: FarInfrared
Emission from Three Local Elliptical Galaxies. A p J , 585:L121-L124, M a r
2003.
E . Tempel, R. Kipper, A . Tamm, M . Gramann, M . Einasto, T. Sepp, and T. Tuvikene.
Friends-of-friends galaxy group finder with membership refinement. Application to
the local Universe. A & A , 588:A14, A p r 2016.
R. C. Thomson. Shell formation in elliptical galaxies. M N R A S , 253:256, Nov 1991.
G. R. Tremblay, F . Combes, J. B . R. Oonk, H . R. Russell, M . A . McDonald, M . Gaspari,
B . Husemann, P. E . J. Nulsen, B . R . McNamara, S. L . Hamer, C . P. O'Dea, S. A .
B a u m , T . A . Davis, M . Donahue, G . M . Voit, A . C . Edge, E . L . Blanton, M . N .
Bremer, E . Bulbul, T. E . Clarke, L . P. David, L . O. V . Edwards, D . Eggerman, A . C.
Fabian, W . Forman, C . Jones, N . Kerman, R. P. Kraft, Y . L i , M . Powell, S. W .
Randall, P. Salome, A . Simionescu, Y . Su, M . Sun, C . M . Urry, A . N . Vantyghem,
B . J . Wilkes, and J. A . ZuHone. A Galaxy-scale Fountain of Cold Molecular Gas
Pumped by a Black Hole. A p J , 865(1):13, Sep 2018.
Bibliography 58
R. B . Tully and N . Trentham. Midlife Crises in Dwarf Galaxies in the N G C 5353/4
Group. A J , 135(4): 1488-1504, A p r 2008.
M . J. L . Turner, A . Abbey, M . Arnaud, M . Balasini, M . Barbera, E . Belsole, P. J. Bennie,
J. P. Bernard, G . F . Bignami, M . Boer, U . Briel, I. Butler, C. Cara, C. Chabaud,
R. Cole, A . Collura, M . Conte, A . Cros, M . Denby, P. Dhez, G . D i Coco, J. Dowson,
P. Ferrando, S. Ghizzardi, F . Gianotti, C . V . Goodall, L . Gretton, R . G . Griffiths,
O. Hainaut, J. F . Hochedez, A . D . Holland, E . Jourdain, E . Kendziorra, A . Lagostina,
R. Laine, N . L a Palombara, M . Lortholary, D . Lumb, P. Marty, S. Molendi, C. Pigot,
E . Poindron, K . A . Pounds, J . N . Reeves, C . Reppin, R. Rothenflug, P. Salvetat,
J. L . Sauvageot, D . Schmitt, S. Sembay, A . D . T . Short, J . Spragg, J . Stephen,
L. Strüder, A . Tiengo, M . Trifoglio, J. Trimmer, S. Vercellone, L . Vigroux, G . Villa,
M . J. Ward, S. Whitehead, and E . Zonca. The European Photon Imaging Camera
on X M M - N e w t o n : The M O S cameras : The M O S cameras. A & A , 365T27-L35,
January 2001.
G. M . Voit and M . Donahue. The Fate of Stellar Mass Loss in Central Cluster Galaxies.
A p J , 738:L24, September 2011.
G . M . Voit and M . Donahue. Cooling Time, Freefall Time, and Precipitation in the
Cores of A C C E P T Galaxy Clusters. A p J , 799:L1, January 2015.
G . M . Voit, C. P. M a , J. Greene, A . Goulding, V . Pandya, M . Donahue, and M . Sun.
A General Precipitation-limited L x - T - R Relation among Early-type Galaxies. A p J ,
853(1):78, Jan 2018.
N . Werner, S. W . Allen, and A . Simionescu. O n the thermodynamic self-similarity of
the nearest, most relaxed, giant ellipticals. M N R A S , 425:2731-2740, October 2012.
N . Werner, J. B . R. Oonk, R. E . A . Canning, S. W . Allen, A . Simionescu, J. Kos, R. J.
van Weeren, A . C . Edge, A . C . Fabian, A . von der Linden, P. E . J . Nulsen, C . S.
Reynolds, and M . Ruszkowski. The Nature of Filamentary Cold Gas in the Core of
the Virgo Cluster. A p J , 767(2):153, A p r 2013.
N . Werner, J . B . R. Oonk, M . Sun, P. E . J. Nulsen, S. W . Allen, R. E . A . Canning,
A . Simionescu, A . Hoffer, T. Connor, M . Donahue, A . C. Edge, A . C. Fabian, A . von
der Linden, C . S. Reynolds, and M . Ruszkowski. The origin of cold gas in giant
elliptical galaxies and its role in fuelling radio-mode A G N feedback. M N R A S , 439:
2291-2306, A p r i l 2014.
R. A . Wood, C. Jones, M . E . Machacek, W . R. Forman, A . Bogdan, F . Andrade-Santos,
R. P. Kraft, A . Paggi, and E . Roediger. The Infall of the Virgo Elliptical Galaxy M60
toward M87 and the Gaseous Structures Produced by Kelvin-Helmholtz Instabilities.
A p J , 847:79, Sep 2017.
L. M . Young, G . J. Bendo, and D . M . Lucero. M i d - to Far-Infrared Emission and Star
Formation in Early-Type Galaxies. A J , 137:3053-3070, Feb 2009.
Bibliography 59
L. M . Young, M . Bureau, T. A . Davis, F . Combes, R. M . McDermid, K . Alatalo, L . Blitz,
M . Bois, F . Bournaud, M . Cappellari, R. L . Davies, P. T. de Zeeuw, E . Emsellem,
S. Khochfar, D . Krajnovic, H . Kuntschner, P.-Y. Lablanche, R. Morganti, T. Naab,
T. Oosterloo, M . Sarzi, N . Scott, P. Serra, and A . - M . Weijmans. The A T L A S 3 D
project - I V . The molecular gas content of early-type galaxies. M N R A S , 414(2):
940-967, Jun 2011.
L. M . Young, N . Scott, P. Serra, K . Alatalo, E . Bayet, L . Blitz, M . Bois, F . Bournaud,
M . Bureau, A . F . Crocker, M . Cappellari, R. L . Davies, T . A . Davis, P. T. de
Zeeuw, P . - A . Due, E . Emsellem, S. Khochfar, D . Krajnovic, H . Kuntschner, R . M .
M c D e r m i d , R. Morganti, T . Naab, T. Oosterloo, M . Sarzi, and A . - M . Weijmans.
The A T L A S 3 D
project - X X V I I . Cold gas and the colours and ages of early-type
galaxies. M N R A S , 444(4):3408-3426, Nov 2014.
M . S. Y u n , N . A . Reddy, and J . J . Condon. Radio Properties of Infrared-selected
Galaxies in the I R A S 2 Jy Sample. A p J , 554:803-822, Jun 2001.
I. Zhuravleva, E . Churazov, A . A . Schekochihin, S. W . Allen, P. Arevalo, A . C .
Fabian, W . R. Forman, J. S. Sanders, A . Simionescu, R. Sunyaev, A . Vikhlinin, and
N . Werner. Turbulent heating in galaxy clusters brightest in X-rays. Nature, 515
(7525):85-87, Nov 2014.