Download IXO as an observatory in the large telescopes era

1
1. Scientific Objectives
1 IXO Science Objectives: The physics of the hot Universe
IXO is an observatory class Astronomy mission that has been conceived to provide direct insight into
some of the most important themes posed by ESA’s Cosmic Vision 2015-2025 science objectives1 :

(Q4.3) The Evolving Violent Universe, by finding the growing massive black holes present in
galaxy centers from their young ages to the present; and understanding how they influence the
history of the formation and growth of their host galaxy through feedback processes.

(Q4.2) The Universe taking shape, by studying how the baryonic component of the Universe
formed large-scale structures in the Universe, finding the still large fraction of those baryons
that are still missing and also understanding how and when the Universe was chemically
enriched by Supernovae

(Q3.3) Matter Under Extreme Conditions, by studying how matter behaves under very strong
gravity field conditions, only attainable around black holes and compact objects, where
General Relativity predicts a number of effects; and also how matter behaves at densities
higher than in atomic nuclei in the interiors of neutron stars.
IXO will also help to provide answers to other Cosmic Vision themes, most notably to (Q4.1) The
Early Universe, by measuring dark matter and dark energy using galaxy clusters. Being an observatory
class mission, IXO will also be able to address a large number of additional problems in contemporary
astrophysics, a few of which are highlighted here: studies of the interstellar medium, star and planet
formation, stellar mass loss and the origin of cosmic rays in Supernovae. Likewise, IXO will provide
responses to a very large fraction of the questions posed by the Astronet Science Vision exercise (de
Zeeuw & Molster 2007), which defines the science goals of European astronomy for the next two
decades. IXO is undoubtedly the single mission that will have the largest impact on the science
objectives of both Cosmic Vision 2015-2025 and Astronet Science Vision.
Figure 1.1. Sensitivity of future
key observatories, along with the
spectrum of a galaxy with strong
star formation and a merging
massive binary black hole, like
NGC 6240 at z=10.
1
Cosmic Vision: Space Science for Europe 2015-2025, ESA BR-247
2
1. Scientific Objectives
1.1 Co-evolution of galaxies and their supermassive black holes
1.1.1 The first supermassive black holes
One of the main themes in extragalactic astronomy for the next decade and beyond will be the
formation and evolution of the first galaxies and black holes. Many future observatories, including
JWST, ALMA, GMT, TMT and E-ELT will intensively observe starlight with a broad range of
redshifts, out to the dawn of the modern Universe when the first galaxies formed. It has, however,
become clear that the properties and evolution of galaxies are intimately linked to the growth of their
central black holes. Any understanding of the formation of galaxies, and their subsequent evolution,
will therefore be incomplete without similarly intensive observations of the accretion light from
supermassive black holes (SMBHs) in galactic nuclei. To make further progress, we need to chart the
formation of typical SMBHs at z>6, and their subsequent growth over cosmic time, which is most
effectively achieved with X-ray observations. The ongoing technological developments in X-ray
optics and instrumentation being undertaken for IXO now bring this within reach, enabling capabilities
fully matched to those expected from flagship observatories at longer wavelengths (Fig. 1.1 the in
previous section).
Black hole accretion and galaxy evolution: a fundamental connection
A major recent development in astrophysics has been the discovery of the relationship in the local
Universe between the properties of galaxy bulges, and dormant black holes in their centers (e.g. the
M- relation between the black hole mass and bulge velocity dispersion; Ferrarese & Merrit 2000;
Gebhardt et al. 2000). These tight relationships represent definitive evidence for the co-evolution of
galaxies and Active Galactic Nuclei (AGN). The remarkable implication of this is that some
consequence of the accretion process on the scale of the black hole event horizon is able to influence
the evolution of an entire galaxy. The main idea is that radiative and mechanical energy from the AGN
regulates both star formation and accretion during periods of galaxy growth.
This kind of black hole driven feedback is thought to be essential in shaping the first galaxies. Current
models propose that mergers of small gas-rich proto-galaxies in deep potential wells at high redshift
drive star formation and black hole growth (in proto-quasar active galaxies) until a luminous quasar
forms. At this point, a black hole driven wind evacuates gas from the nascent galaxy, limiting
additional star formation and further black hole growth (Silk & Rees 1998; Fig. 1). Further episodes of
merger-driven star formation, accretion, and feedback are expected to proceed through cosmic time.
This provides a plausible origin for the M- relation (e.g. King 2003). It also explains many
outstanding problems in galaxy evolution (e.g. Croton et al. 2005; Hopkins et al. 2006). Despite the
intense current interest in this topic, and its great importance, the physical processes are unclear, and
direct evidence for this kind of AGN feedback is scarce. There are also many possible scenarios for
the formation and growth of the earliest SMBH. This topic is therefore certain to remain one of the
most active topics in astrophysics, not least because an X-ray observatory with the deep survey
capabilities of IXO is required for us to obtain a full picture of the role of black holes in the Universe.
3
1. Scientific Objectives
Figure 2.2. Formation of a high-redshift
quasar from hierarchical galaxy mergers as
simulated by Li et al. (2007). Color shows
gas temperature, and intensity shows gas
density. Black dots represent black holes.
Small, gas-rich galaxies merge in the deepest
potential wells at high redshift, promoting
star formation and black hole growth. At z~7
to z~5 a luminous quasar forms, associated
with the most massive black hole. It drives a
wind (yellow) that evacuates gas from the
nascent galaxy, terminating star formation
and self-regulating accretion in a process
which has become known as “feedback”.
Black hole growth in the early Universe
The very first stars formed in primordial structures where gravity was able to overpower the pressure
of the ambient baryons, a few hundred million years after the Big Bang (z~20). The first seed black
holes were left behind as remnants of the most massive stars. The first galaxies, hosting these first
black holes in their cores, were responsible for reionizing the Universe by z~10, as shown by WMAP
(Dunkley et al. 2009). Still, the highest redshift galaxies and quasars currently known are all in the
range z=6–8. To understand the inner workings of the first luminous sources we need to bridge the gap
between the few known sources at this redshift, and the information we can extract from the
microwave background.
The known AGN population at z=6-7 currently consists of luminous optical quasars (e.g. Fan et al.
2003). Growing the extremely massive black holes required in <1 Gyr represents a challenge for
theoretical models, because it requires Eddington-limited accretion over many doubling times. Gasdynamical cosmological simulations are nevertheless able to produce quasars with ~10 9 M at z=6.5
through a rapid sequence of mergers in small groups of proto-galaxies (Li et al. 2007; Fig. 1.2). The
growth is likely to proceed in a self-regulated manner owing to feedback with the progenitor host, with
a period of intense star formation and obscured accretion preceding the optically bright quasar phase.
The complex physics involved in such a scenario is, however, poorly understood. Furthermore,
evidence for widespread merger-driven AGN activity and feedback at high redshift is scarce. And
alternative SMBH formation and growth scenarios are possible. For example, they may form via
direct, monolithic collapse of hot, dense gas clumps (Bromm & Loeb 2003; Begelman et al. 2006) or
via “quasistars” (Begelman et al. 2008).
It must also be borne in mind that luminous, optical QSOs, hosting among the most massive black
holes (>109 M) in the Universe, are extremely rare. Typical AGN, which are of lower luminosity and
often obscured, remain largely undiscovered. Uncovering such objects at z=6-7 (and even higher
redshifts) holds the key to our understanding of this crucial phase in the development of the Universe.
It is very likely that SMBHs as massive as 106 M, possibly hosted by vigorously star forming
galaxies, existed as early as z=10–11. X-ray observations offer a unique tool to discover and study the
accretion light from moderate luminosity AGN at z=6-11, which are rendered invisible in other
wavebands due to intergalactic absorption and dilution by their host galaxy. Testing the various
competing scenarios for the evolution of early SMBH will be rendered possible by IXO, which will
reveal the AGN population at the highest redshifts for the first time.
4
1. Scientific Objectives
IXO: Breaking through to typical AGN at “first light”
The highest redshift QSOs known have been discovered in wide-field optical surveys (e.g. SDSS; Fan
et al. 2001). These are fascinating sources, but it is crucial to remember that they are among the most
extreme and unusual objects in the Universe. Large area optical surveys will be continued with, e.g.
Pan-STARRS, LSST, VISTA, and perhaps JDEM/EUCLID, which will discover many more high z
QSOs. The optical surveys are, however, fundamentally limited to objects in which the AGN
outshines the galaxy. For this reason, X-ray observations can probe much lower bolometric
luminosities than the optical. Current deep X-ray surveys probe factors of 100-1000 fainter down the
luminosity function than SDSS, at or around L*, where the bulk of the accretion power is produced.
We furthermore expect the majority of AGN at high redshift to be heavily obscured by gas and dust,
where the accretion light is rendered invisible in the optical but detectable at X-ray energies. X-ray
observations are thus absolutely essential in both the discovery and characterization of typical
accreting black holes at high redshift.
Figure 1.3. Simulations of 1 Ms and 100 ks IXO observations of the Chandra Deep Field South, compared to
the Chandra 2 Ms image. The IXO simulation is based on the Chandra image, with additional random sources
added with fluxes below the sensitivity limits of the Chandra including both AGN and normal galaxies. IXO
reaches the same depth as Chandra in ~1/20 th of the exposure time, meaning it can both reach greater depths
and cover much larger areas, essential to uncover large samples of the earliest supermassive black holes (z>7).
The current deepest surveys with Chandra reach great depths at the central aim point. These have
yielded a handful of AGN candidates at z>5 (Luo et al 2010), but none is yet confirmed at z>6. To
harvest significant samples of moderate luminosity AGN at z=7-10, however, we need to reach this
kind of depth over much larger areas. IXO's combination of huge collecting area and excellent spatial
resolution over a large field of view fulfils this requirement (Fig. 1.3). These observations will have
enormous constraining power. Current semi-analytic models of galaxy formation in the early Universe
predict vastly different amounts of black hole growth in the early universe (Fig 1.4 e.g. Rhook &
Haehnelt 2008; Salvaterra et al 2007; Marulli et al. 2008). Extrapolations of the current best X-ray
luminosity functions (XLFs) are almost as uncertain (Brusa et al. 2008; Aird et al. 2008; Silverman et
al. 2008; Ebrero et al. 2009). Taking the best (median) estimates from these, a typical “multi-layered”
IXO survey with ultradeep (2x1-2 Ms) deep (12x300 ks) and wide (24x100 ks) components will yield
several hundred X-ray selected AGN at z~6, and as much as a few 10s at z~10, depending on model
extrapolations. Additional very high redshift AGN will be identifiable via serendipitous observations
with the WFI, which will yield a rich archive, but dedicated surveys are strongly preferable as they
will target the same areas as the complementary multiwavelength facilities (Fig. 1.1). These are
necessary for the identification and redshift determination of faint X-ray sources pinpointed to high
accuracy with IXO’s excellent angular resolution. Facilities like JWST and ALMA will also yield the
host galaxy properties of these early AGN (e.g. stellar and gas masses, star formation rates), which
will be crucial in discriminating between various SMBH formation models. For example, mergerdriven models predict large star formation rates and substantial stellar masses in early AGN hosts.
5
1. Scientific Objectives
Monolithic formation models predict neither, as the black hole forms first without fragmentation of the
collapsing gas cloud. These considerations again emphasize the high degree of complementarity
between IXO and the next generation of observatories across the electromagnetic spectrum.
Figure 1.4. Predicted number counts
at z > 6 for various semi-analytic
models of galaxy formation. There is
an extraordinary range of predictions,
which
will
be
distinguished
straightforwardly with IXO. Revealing
the entire AGN population in X-rays,
and characterizing the host galaxy
populations with the next generation
of longer-wavelength facilities (Fig.
1.1) will be of fundamental importance
to our understating of the early
Universe.
1.1.2 Obscured growth of supermassive black holes
The tight correlation between the mass of galactic bulge and that of supermassive black holes
(SMBHs) in their centre found in the local universe indicates that the star formation and mass
accretion co-evolved throughout the history of the universe. Since Active Galactic Nuclei (AGNs) are
the very phenomena where black holes grow by mass accretion, we need to uncover all populations of
AGN over a broad range of redshifts, luminosities, and obscuration, to understand the whole growth
history of supermassive black holes and their relation to galaxy formation. Theories suggest that a
significant portion of SMBH growth in massive galaxies occurred during a phase of heavy
obscuration, accompanied by intense star formation (e.g., Hopkins et al. 2006). Thus, it is a key issue
to reveal the evolution of obscured AGNs, in particular at redshift range of 1-3, where both mass
accretion and star formation had peak activities in the cosmic history.
Due to difficulties of detecting heavily obscured AGNs (including "Compton thick" AGNs with NH >
1024 cm-2), we are far from a complete census of the AGN population, however, at the whole redshift
range even in the local universe. In these objects, the direct emission in the UV, optical, and near IR
bands as well as at E<10 keV from the nucleus is blocked by obscuring matter, making it difficult to
probe the central engine. Even in the deepest Chandra and XMM-Newton surveys now available,
~50% of the X-ray background (XRB) above 6-8 keV is still left unresolved (Worsley et al. 2005),
where Compton thick AGNs likely largely contribute. Although indirect estimate on the number
density of Compton thick can be made in the framework of population synthesis models (e.g., Gilli et
al. 2007), such discussion is highly uncertain because their contribution to the XRB is inevitably
coupled with parameters such as the average shape of broad band spectra of Compton thin AGNs,
contribution of minor populations like blazars, and precise (<10%) level of the absolute intensity of
the XRB. More importantly, it is not clear if Compton thick AGNs follow the same cosmological
evolution as found for Compton thin AGNs showing "down-sizing" behavior. The only way to reliably
determine the evolution of Compton thick AGNs is to directly detect them with much increased
sensitivities than current observatories.
Many observations suggest the presence of a large (but uncertain) number of Compton thick AGNs in
the local universe. Evidence for heavily obscured AGNs is found from optical narrow emission line
galaxies (Risaliti et al. 1999) and from infrared selected galaxies (e.g., Maiolino et al. 2001, Imanishi
et al. 2007). Their number density is estimated to be comparable to that of known, Compton thin type-
6
1. Scientific Objectives
2 Seyfert galaxies. From hard X-ray (>10 keV) all sky surveys with Swift and INTEGRAL, Suzaku
has discovered a population of deeply "buried" AGN with very small fractions of scattered lights
indicating a geometrically thick torus around the SMBH (Ueda et al. 2007). This implies that many
similar objects, if observed in edge-on geometry, could be missed in the current hard X-ray surveys
due to the suppression of transmitted emission as they become heavily Compton thick (NH > a few
1024 cm-2). This bias may explain the apparently very small fraction of Compton thick AGNs in the
Swift/BAT sample (Tueller et al. 2009).
At higher redshifts (z~2), it has been suggested from deep multi-wavelengths surveys that Comptonthick AGN at z~2 may be hiding among infrared bright, optically faint galaxies, although each object
is generally too faint in X-rays to detect individually even with deepest Chandra exposures (e.g. Daddi
et al. 2007; Alexander et al. 2008; Fiore et al. 2008). If one assumes that all members in the sample
have the same hard X-ray flux, their average corresponds to an intrinsic X-ray luminosity > 1043 erg s1
, and the estimated number density is even higher than that of Compton-thin AGNs at the same
luminosity and redshift range. If there is a large population of such objects at cosmological redshifts,
they could make a major contribution to the total accretion power (Fabian & Iwasawa 1999). This can
change our view of the accretion history of the universe established from Compton-thin AGNs, where
the mass function of SMBHs in the local universe can be well explained by standard accretion mode
based on Soltan's argument (Marconi et al. 2004). The inferred very high number density of Comptonthick AGNs at z~2 is qualitatively consistent with the evolution of absorbed fraction seen in Compton
thin AGNs (i.e, more obscuration at higher redshifts; La Franca et al. 2005, Hasinger 2008). However,
simple AGN unification scheme may not hold since the evolutionary phase of Compton thick AGNs in
galaxy evolution could be different from that of Compton thin AGNs; models predict that Comptonthick AGNs correspond to the stage where the AGN blows out gas from the galaxy and terminates star
formation (Hopkins et al. 2006).
While X-ray selection has provided the most robust AGN samples to date, finding the most obscured
objects has proved difficult with current X-ray missions. Mid-IR selection is promising, but has not
yet yielded samples that are both reliable and complete, due to the contamination from star burst
activities in separating AGN components. Future hard X-ray (>10 keV) imaging (NuSTAR, ASTROH) will provide a step forward in revealing AGN with column densities of NH ~ a few 1024 cm-2. At the
highest column densities, even the 10-40 keV light is suppressed (by a factor ~10 at NH =1025 cm-2),
leaving the AGN visible only in scattered X-rays. The spectral sensitivity of IXO in the 2-10 keV band
will reveal the telltale intense iron K emission characteristic of a Compton reflection dominated source
(Figure 1.5) and can be combined simultaneously with hard X-ray data of unprecedented sensitivity.
From deep surveys of IXO, it is expected that more than 50%-80% of the 10-40 keV XRB can be
resolved into discrete AGNs including Compton thick ones (Figure 1.6). Identification and
investigation of the nature of these objects (e.g., star formation rate) can be made from longer
wavelengths dataset that will become available by future observatories, such as ALMA, JWST, EELT, and TMT (Figure 1.1). This study is a key step for understanding the co-evolution of galaxies
and SMBHs when their activities are the most significant.
7
1. Scientific Objectives
Figure 1.5. Left: Simulated spectra of the z=3.7 Compton thick AGN CDFS-202 (Norman et al. 2002), in a 1Ms
exposure with the IXO WFI; Parameters are based on the deep XMM observation of the CDF-S (PI: Comastri)
Right: the characteristic X-ray BAL signatures of the fast (~0.2c) outflow in APM 08279+5255 (Chartas et al.
2002), which may be in the “blowout phase”, as seen by the IXO/XMS calorimeter in just 20ks.
Figure 1.6. Left: Estimated sensitivity of IXO in the 10-40 keV band as a function of exposure, which saturates
at the 30-beams confusion limit. The two curves correspond to the performance of requirement (green) and
goal (blue.) Right: the resolved fraction of the 10-40 keV XRB as a function of sensitivity, and its expectation
from IXO, based on the model by Gilli et al. (2007).
1.1.3 Cosmic feedback from supermassive black
An extraordinary recent development in astrophysics was the discovery of the fossil relationship
between central black hole mass and the stellar mass of galactic bulges. The physical process
underpinning this relationship has become known as feedback. The Chandra X-ray Observatory was
instrumental in realizing the physical basis for feedback, by demonstrating a tight coupling between
the energy released by supermassive black holes and the gaseous structures surrounding them. A great
leap forward in X-ray collecting area and spectral resolution is now required to address the following
question: How did feedback from black holes influence the growth of structure?
Feedback and galaxy formation
Every massive galaxy appears to have a massive black hole at its center whose mass is about 0.2% of
the mass of the galaxy's bulge (Tremaine et al. 2002). It is now widely considered that the black hole
may have brought about this correlation by regulating the amount of gas available for star formation in
the galaxy. Massive black holes thereby have a profound influence on the evolution of galaxies, and
possibly on their formation. Several puzzling aspects of galaxy formation, including the early
quenching of star formation in galactic bulges and the galaxy mass function at both high and low ends,
have been attributed to black hole “feedback” (Croton et al. 2006).
8
1. Scientific Objectives
In size, a black hole is to a galaxy roughly as a person is to the Earth. Something very small is
determining the growth of something very large. This is possible because the gravitational potential
energy acquired by an object approaching a black hole is a million times larger than the energy of an
object orbiting in the potential of a typical galaxy. As a black hole grows to 0.2% of the bulge mass
through accreting matter, it releases nearly 100 times the gravitational binding energy of its host
galaxy.
There is no question that a growing black hole could drastically affect its host galaxy. Whether and
how it does so, however, is an open question that depends on how much of the energy released
actually interacts with the matter in the galaxy. If the energy is in electromagnetic radiation and the
matter largely stars, then very little interaction is expected. If the matter consists of gas, perhaps with
embedded dust, the radiative output of the black hole can both heat the gas, and drive it via radiation
pressure. Alternatively, if significant AGN power emerges in winds or jets, mechanical heating and
pressure provide the link. Either form of interaction can be sufficiently strong that gas can be driven
out of the galaxy entirely (Silk & Rees 1998).
The radiative form of feedback is most effective when the black hole is accreting close to its
Eddington limit. The mechanical form associated with jets, on the other hand, operates at rates below
the Eddington limit. X-ray observations are essential for studying both forms of feedback. The
mechanical forms of feedback rely on dynamical (ram) pressure to accelerate gas to high speeds (Fig.
1.7). If this gas is initially of moderate temperature, the interaction will shock it to high temperatures
where it can only be detected in X-rays. Since the efficiency of ram pressure acceleration is roughly
proportional to the volume filling factor of the accelerated gas, most of the energy is probably
absorbed by the hot component of the interstellar medium in any case. Gas accelerated by radiation
pressure or radiative heating is likely to be cold and dusty. The interaction is therefore much more
difficult to observe directly. X-ray and far infrared emission can emerge from the inner regions where
the interaction occurs, revealing the Active Galactic Nucleus (AGN) itself.
Figure 1.7. Left: X-ray emission (blue), radio emission (red) superposed on an optical image of M87. The Xray structure (shocks, bubbles) was induced by ~10 58 erg outburst that began 10 Myr ago (Forman et al. 2005).
The persistence of the delicate, straight-edge X-ray feature indicates a lack of strong turbulence. We expect IXO
to reveal ordered velocity structure. The image is 50 kpc on a side. Right: X-ray emission (blue), 320 MHz
radio emission (red) superposed on an HST image of the z=0.21 cluster MS0735.6+7421 (McNamara et al.
2005). The image is 700 kpc on a side. Giant, cavities, each 200 kpc (1 arcmin) in diameter were excavated by
the AGN. The mechanical energy is reliably measured in X-rays by multiplying the gas pressure by the volume
of the cavities, and by the properties of the surrounding shock fronts. With a mechanical energy of 10 62 erg,
MS0735 is the most energetic AGN known. This figure shows that AGN can affect structures on galaxy scales of
tens of kpc and on cluster-wide scales spanning hundreds of kpc in MS0735.
9
1. Scientific Objectives
Radiative acceleration appears to be particularly dramatic in the outflows from some luminous
quasars. UV observations indicate that outflows reaching 0.1-0.4c may be present in most quasars. Xray observations are required to determine the total column density and hence the kinetic energy flux.
Current work on a small number of objects implies that this can be comparable to the radiative
luminosity (Fig. 1.8). To diagnose the energetics of quasars we need large samples of quasars,
comparing them with the less energetic (but still substantial) outflows from lower luminosity AGN,
which can reach speeds of several thousand km/s.
Figure 1.8. Top: Suzaku X-ray
spec-trum of luminous quasar
PDS456 at z=0.184.
The
strong iron absorption lines at
9 keV in the rest frame indicate
an outflow at 0.3c and a
mechanical power of 1047 ergs-1
(Reeves et al. 2009).
Bottom: 20 ksec simulation of
an IXO spectrum of the
absorption structure of the
mildly relativistic outflow seen
by Chandra in the BAL quasar
APM08279 at z=3.91.
IXO’s huge spectroscopic throughput will extend this to redshifts z=1-3, where the majority of galaxy
growth is occurring. IXO will be sensitive to all ionization states from Fe I – Fe XXVI, allowing us to
study how feedback affects all phases of interstellar and intergalactic gas, from million-degree
collisionally ionized plasmas to ten-thousand degree photoionized clouds. These measurements will
probe over 10 decades in radial scale, from the inner accretion flow where the outflows are generated
to the halos of galaxies and clusters, where the outflows deposit their energy.
Current models of galaxy evolution predict that merger-induced star formation and AGN activity
proceeds under heavy obscuration, building the galaxy’s bulge and black hole before the AGN blows
out all of the gas and terminates star formation. Following this, an unobscured QSO is briefly
revealed before the galaxy becomes much less active as a massive red elliptical. The complete census
of AGN enabled by IXO will pinpoint galaxies whose black holes are undergoing all these phases of
evolution, and crucially the heavily obscured “Compton thick” phase predicted during the quenching
10
1. Scientific Objectives
of star formation. Observations of galaxies in the X-ray band are the only way to select accreting
black holes in an unbiased fashion, and to probe the inner workings of AGN near the black hole’s
gravitational radius.
Thermal regulation of gas in bulges and cluster cores
Mechanical feedback dominates in galaxies, groups, and clusters at late times, as shown by X-ray
observations of gas in the bulges of massive galaxies and the cores of galaxy clusters (eg., Fig. 1.7).
The energy transfer process is surprisingly subtle. The radiative cooling time of the hot gas in these
regions is often much shorter than the age of the system, so that without any additional heating, the gas
would cool and flow into the center. For giant ellipticals the resulting mass cooling rates would be of
order 1 solar mass per year. At the centres of clusters and groups, cooling rates range between a few
to thousands of M per year. Spectroscopic evidence from Chandra and XMM show that some cooling
occurs, but not to the extent predicted by simple cooling (Peterson & Fabian 2006). Limits on cool
gas and star formation rates confirm this. Mechanical power from the central AGN acting through jets
must be compensating for the energy lost by cooling across scales of tens to hundreds of kpc
(McNamara & Nulsen 2007).
The gross energetics of AGN feedback in galaxies and clusters are reasonably well established (Fig.
1.9). Remarkably, relatively weak radio sources at the centers of clusters often have mechanical
power comparable to the output of a quasar, which is sufficient to prevent hot atmospheres from
cooling (McNamara & Nulsen 2007). The coupling between the mechanical power and the
surrounding medium are, however, poorly understood. Moreover, it is extremely hard to understand
how a fine balance can be established and maintained.
Figure 1.9. This Figure needs a caption
The heat source – the black hole – is roughly the size of the Solar System, yet the heating rate must be
tuned to conditions operating over scales 10 decades larger. The short radiative cooling time of the
gas means that the feedback must be more or less continuous. How the jet power, which is highly
collimated to begin with, is isotropically spread to the surrounding gas is not clear. The obvious signs
of heating include bubbles blown in the intracluster gas by the jets (Figs. 1.7, 1.10) and nearly quasispherical ripples in the X-ray emission that are interpreted as sound waves and weak shocks. Future
low-frequency radio observations of the bubbles and cavities are of great importance in determining
the scale of the energy input. The disturbances found in the hot gas carry enough energy flux to offset
cooling, but the microphysics of how such energy is dissipated in the gas is not understood.
The persistence of steep abundance gradients in the cluster gas, imprinted by supernovae in the central
galaxy means that the feedback is gentle, in the sense that it does not rely on violent shock heating or
11
1. Scientific Objectives
supersonic turbulence. Long filaments of optical line-emitting gas in some objects suggest low levels
of turbulence. Yet the continuous streams of radio bubbles made by the jets, the movement of
member galaxies and occasional infall of subclusters must make for a complex velocity field.
With high resolution imaging and moderate resolution spectroscopy (Fig. 1.11), the Chandra and
XMM-Newton observatories have established AGN feedback as a fundamental astrophysical process in
nature. However, the dynamics of these powerful outflows are not understood.
Figure 1.10. Right: How did feedback from black holes
influence galaxy growth? Chandra X-ray observations of
Perseus (left; Fabian et al. 2003) and other nearby clusters
have revealed the indelible imprint of the AGN on the hot gas
in the core. Radiative and mechanical heating and pressure
from black holes has a profound influence not only on the hot
baryons, but on the evolution of all galaxies whether or not
they are in clusters.
Figure 1.11. Left: XMM-Newton
reflection grating spectrometer
spectrum of the core of the
Centaurus cluster (Sanders et al.
2008). This part of the spectrum
shows a rich set of emission lines
from oxygen, neon, and iron. The
ionization states of iron show the
temperature structure of the gas
ranging from 0.4 keV to 4 keV.
This demands a leap in spectral resolution by one to two orders of magnitude above that of Chandra
and XMM-Newton. The spectral resolution and sensitivity of the next generation X-ray observatory,
such as IXO, is needed to understand how the bulk kinetic energy is converted to heat. Its capabilities
are essential in order to measure and map the gas velocity to an accuracy of ten km/s, revealing how
the mechanical energy is spread and dissipated (Fig. 1.12). From accurate measurements of line
profiles and from the variations of the line centroid over the image it is possible to deduce the
characteristic spatial scales and the velocity amplitude of large (> kpc) turbulent eddies, while the total
width of the line provides a measure of the total kinetic energy stored in the stochastic gas motions at
all spatial scales. Such data will provide crucial insight into the ICM heating mechanisms.
Observations of the kinematics of the hot gas phase, which contains the bulk of the gaseous mass, and
absorbs the bulk of the mechanical energy in massive elliptical galaxies, are only possible at X-ray
wavelengths.
12
1. Scientific Objectives
Figure 1.12. Below: Simulated high-resolution X-ray spectra from the shells and X-ray cavities in the Perseus
cluster (see Fig. 3), demonstrating the power of imaging spectroscopy for AGN feedback studies. The right
panel shows the X-ray image and the chosen cut (spectral slit, slicing through both cavities), while the left shows
the spectrum of the K-alpha lines from Fe XXV and Fe XXVI (both lines are multiplets), as would be observed by
the IXO micro calorimeter spectrograph in an exposure of 250,000 seconds. At the location of the cavities
(y=10-15 and y=25-30), each of the lines splits into three components: The front wall of the cavity (blueshifted), the rear wall of the cavity (red-shifted), and a rest-frame component from fore- and background cluster
emission. The expansion velocity and thus the age of the cavity can easily be determined from the red- and blue
shifts of the lines, allowing a precise determination of the jet power. The cluster/radio galaxy model used for the
spectral simulations is the result of direct hydrodynamic simulations of jets in galaxy clusters with parameters
appropriate for Perseus (jet power 1045 erg s-1; Heinz et al. 2009, in preparation).
See www.astro.wisc.edu/~heinzs/perseus for a movie.
1.2 Large scale structure and the creation of chemical elements
1.2.1 Missing baryons and the Warm-Hot Intergalactic Medium
Ordinary matter (baryons) represents 4.6% of the total mass/energy density of the Universe but less
than 10% of this matter appears in collapsed objects (stars, galaxies, groups; Fukugita & Peebles
2004). Theory predicts that most of the baryons reside in vast unvirialized filaments that connect
galaxy groups and clusters (the “Cosmic Web”; Fig. 1.13 right). After reionization, the dominant
heating mechanism is through the shocks that develop when large-scale density waves collapse in the
dark matter. As large-scale structure becomes more pronounced with cosmological time, the gas is
increasingly shock-heated, reaching temperatures of 105. 5-107 K for z < 1 (Fig. 1.13 left). Additional
heating occurs through star-formation driven galactic winds and AGN, processes that pollute the
surroundings with metals.
13
1. Scientific Objectives
Figure 1.13. Left: The differential gas mass fraction at as a function of temperature at low redshift for the
ΛCDM cosmological simulation of Cen & Ostriker (2006). This distribution is sensitive to the presence of
galactic superwinds (solid red line; dashed line is without superwinds). The ions with the strongest resonance
lines in the 105-107 K range are shown, and except for OVI (UV line; 1035 Å), the other lines lie in the X-ray
band. Right: The density distribution of baryons at low redshift from the same simulations. Most of the mass of
the WHIM lies within the filaments that connect the higher density regions.
Ly  studies and OVI absorption line studies detect warm baryons, but ~50% of the baryons remain
unaccounted for (Danforth & Shull 2008). These “missing baryons” can only be observed through Xray studies. Therefore, a basic goal is to
Determine if the missing baryons exist in the predicted hot phase
In the standard cosmology, chemical enrichment of the IGM occurs through galactic superwinds, a
powerful feedback mechanism that also heats the gas. The shocks and superwinds leave distinctive
features on absorption lines, such as double-lines (for a line of sight passing through a galactic
superwind shell) and turbulent broadening. This feedback mechanism not only extends the cross
section of the metal-enhanced regions, its effects are ion-dependent. These observational diagnostics
allow us to pursue our second science goal:
Test the large-scale structure and galactic superwind heating of the Cosmic Web
The extent of the superwinds and the elemental mixing can be determined by studying the spatial
relationship between hot gas seen through X-ray absorption and the location of galaxies (Stocke et al.
2006). By studying the same sight lines with X-rays and UV-optical bands, we will discover the
relationship of all temperature components to the galactic environment. Therefore, X-ray studies will
Measure the extent of galactic superwinds and the chemical mixing process
Finally, the baryons lie in Cosmic Web filaments extending between groups and clusters at 1-2 Rvirial,
as seen in Fig.1.13. Current surface brightness measurements rarely extend beyond 0.7Rvirial, yet they
already suggest examples of the expected phenomenon (e.g., Werner et al. 2008). Studies of these
structures become feasible when the limiting X-ray surface brightness can be lowered by an order of
magnitude. This improvement is now possible, so our last goal is achievable:
Measure the connections of Cosmic Web filaments to galaxy groups and clusters
The first three goals are achievable by measuring the He-like and H-like X-ray resonance lines of
carbon (C V, C VI), nitrogen (N VI, N VII) and oxygen (O VII, OV III) toward background AGNs
(other ions may be detected in a few cases, such as Ne IX, Ne X, Fe XVII, and Fe XVIII). Existing
14
1. Scientific Objectives
measurements of intergalactic OVII and OVIII are near current instrumental detection thresholds and
therefore need confirmation (Nicastro et al. 2005; Kaastra et al. 2006; Bregman 2007; Buote et al.
2009) but the adjacent ion, intergalactic OVI, is detected in the UV along many sightlines and there
are clear detections of OVII and OVIII within the Local Group. A conservative estimate of the
equivalent width distribution, dN/dz, is obtained from models normalized to the OVI measurements
(Cen and Fang 2006, Fig. 1.14; the quality of the OVI normalization will improve significantly with
upcoming COS observations). These show that we need an order of magnitude improvement over
current sensitivities to conduct an X-ray survey of intergalactic absorption lines from the above
elements. In addition, to study the velocity structures of lines, we need resolution that approaches the
Doppler width of a line, typically 50-100 km/s, since superwinds from galaxies occur near the escape
velocity, 200-1000 km/s.
The International X-ray Observatory (IXO) has two instruments that can detect these absorption lines,
although the grating spectrometer has a sensitivity advantage and it has the ability to study velocity
structure. The sensitivity to line detection is 15 times better than with XMM or Chandra. With the
baseline design, the grating spectrometer can study the brightest half dozen sources in 2 Msec, while a
sample of the 30 most suitable background AGNs can be studied in 18 Msec, a multi-year project.
From this data set, we will obtain at least 100 absorption line systems, each having one or more lines
from the He-like and H-like species of O, C, and N, which cover the temperature range 105-107 K.
Simulations show that the OVII line will be the strongest, but that multiple lines will be detected in
about 80% of the systems. For each line, we obtain the column density velocity and the velocity
width. This data set will revolutionize the field by answering the four fundamental questions in the
study of the missing baryons.
Figure 1.14. The differential number of absorbers
as a function of equivalent width for OVI (λ1035),
OVII (Kα), and OVIII (Kα), based on the model of
Cen & Fang (2006); the OVI data are from
Danforth and Shull (2005). Current claims of
OVII and OVIII absorption are well above
predictions and would indicate a temperature
distribution significantly different than these
models. The OVII and OVIII absorption lines
predicted from the models, such as those in the
range 2-15 mÅ, are not accessible to Chandra or
XMM, but can be measured with IXO.
Figure 1.15. Simulations for IXO, with the
baseline configuration, of the absorption by the
Cosmic Web seen against a background AGN
with a 0.5-2 keV flux of 510-11 erg cm-2 s-1 and
an exposure time of 600 ksec (zabs = 0.1); this is
the 16th brightest AGN in a sample of 30. There
is a multi-temperature gas (as given by Figure
1) with a superwind, producing a double-lined
configuration with a separation of 500 km/s.
For an AGN at an emission redshift of 0.3,
calculations predict one absorption system with
the strength shown. The lines shown here
would be the signature of a multi-temperature
medium with a temperature range of nearly an
order of magnitude; additional lines are also
detected (NVI, NVII).
15
1. Scientific Objectives
1.2.2 Cluster physics and evolution
The formation and evolution of the large-scale structure of the Universe is a central issue of
cosmology. Recent observational progress, showing that 95% of the total mass-energy of the Universe
is contained in cold dark matter and dark energy, has allowed us to robustly define the cosmological
framework in which structures form. Much progress has also been made in reconstructing the
evolution of the dark matter distribution from its initial density fluctuations. In contrast, we still do not
understand the evolution of the baryonic ‘visible’ component of the Universe, collected in the dark
matter potentials. Present observations, as well as theoretical work, indicate that this is fundamentally
a multi-scale problem, involving a complex interplay between gravitational and poorly understood
non-gravitational processes. Galaxy formation depends on their large scale environment and on the
physical and chemical properties of the intergalactic gas from which they form, which in turn is
affected by galaxy feedback through energy released from star explosion and active nuclei accreting
matter from their environment. Due to the complex behaviour of the baryonic matter, progress has
been driven largely by observations and requires us to study simultaneously the evolution of the hot
and cold components of the Universe.
Galaxy clusters, the largest collapsed structures defining the nodes of the cosmic web, are privileged
sites to comprehend such complex structure formation physics. Over 80% of their mass (up to 1015
MO) resides in the form of dark matter. The remaining mass is composed of baryons, most of which
(about 85%) is a diffuse, hot T > 107 K plasma (the intra-cluster medium, ICM) that radiates primarily
in the X-ray band. Thus in galaxy clusters, through the radiation from the hot gas and the galaxies, we
can observe and study the interplay between the hot and cold components of the baryonic matter and
the dark matter. X-ray observations of the evolving cluster population provide a unique opportunity to
address such open and fundamental questions as:
 How do hot diffuse baryons dynamically evolve in dark matter potentials?
 How and when was the excess energy that we observe in the intergalactic medium generated?
Detailed studies of clusters are currently limited to the relatively nearby Universe (z<0.5). We
need to study the thermo-dynamic and chemical properties of the first low mass clusters (few
1013 MO) emerging at z ~ 2 and directly trace their evolution into today's massive clusters.
This requires the high throughput of IXO, combined with its high spectral and spatial
resolution allowing us to map the ICM density, temperature and metal content. X-ray
observations at high spectral resolution will also open completely new vistas in discovery
space by directly probing the dynamics of the hot gas by mapping the velocity field and
turbulence.
How do hot baryons dynamically evolve in dark matter potentials?
Clusters grow via accretion of dark and luminous matter along filaments and the merger of smaller
clusters and groups. X-ray observations show that many present epoch clusters are indeed not relaxed
systems, but are scarred by shock fronts and contact discontinuities (Markevitch & Vikhlinin 2007),
and that the fraction of un-relaxed clusters likely increases with redshift. Although the gas evolves in
concert with the dark matter potential, this gravitational assembly process is complex, as illustrated by
the temporary separations of dark and X-ray luminous matter in massive merging clusters such as the
“Bullet Cluster'' (Clowe et al. 2006). In addition to the X-ray emitting hot gas, the relativistic plasma
seen through radio synchrotron emission in merging clusters is an important ICM component with at
present few observational constraints.
There are important questions to be answered; both to understand the complete story of galaxy and
cluster formation from first principles and, through a better understanding of cluster physics, to
increase the reliability of the constraints on cosmological models derived from cluster observations
16
1. Scientific Objectives
(see Section 1.2.3). These include: (1) How is the gravitational energy that is released during cluster
hierarchical formation dissipated in the intra-cluster gas, thus heating the ICM, generating gas
turbulence, and producing significant bulk motions? (2) What is the origin and acceleration
mechanism of the relativistic particles observed in the ICM? (3) What is the total level of non-thermal
pressure support, which should be accounted for in the cluster mass measurements, and how does it
evolve with time? To answer these questions, we need to map velocities and turbulence which requires
more than an order of magnitude improvement in spectral resolution, as compared to CCD resolution,
while keeping good imaging capabilities.
Figure 1.16 Left: The 6 keV Fe line region of three simulated IXO calorimeter spectra for a 1 arcmin 2 region in
the very faint shock of the A3667 (z=0.055) merging cluster. The exposure time of the simulations is 200 ksec.
The three spectra correspond to different levels of turbulence, with line widths of 100, 300, and 500 km\s. Since
this cluster is undergoing a merger, we also expect to see line shifts due to gas bulk motions. Right: Zoom on the
Fe line in the overall spectrum (200 ksec observation) for a z=1 cluster of luminosity 3x10 44 ergs/s for two levels
of turbulence.
High-resolution X-ray spectral imaging with IXO will determine the sub-cluster velocities and
directions of motions by combining redshifts measured from X-ray spectra (which give relative lineof-sight velocities) and total sub-cluster velocities deduced from temperature and density jumps across
merger shocks or cold fronts (Markevitch & Vikhlinin 2007). These measurements combined with
high quality lensing observations from instruments such as the LSST or EUCLID will probe how the
hot gas reacts in the evolving dark matter potential. X-ray line width measurements will allow the
level of gas turbulence to be mapped in detail for the first time. As an example, Fig. 1.16 shows that
the 2.5 eV resolution of the IXO calorimeter can distinguish line widths of 100, 300, and 500 km/s in a
small (1 arcmin2) region of the very faint shock of the merging nearby cluster A3667. For a high
redshift cluster (with a luminosity of a few 1044 erg/s at z = 1), the line width could still be measured
to an accuracy of ~100 km/s in a 200 ksec exposure, and more precisely if more time is invested.
IXO is providing the high sensitivity and spatial resolution in the hard energy band ([10-40] keV) to
detect and map the inverse Compton emission that has so far not even clearly been detected. This
promises unique information on the energy density of the relativistic particles, and when combined
with next generation radio observatories like SKA, would probe the history of magnetic fields in
clusters. Capabilities like those of IXO are needed to understand these observationally elusive, but
important components of the ICM.
17
1. Scientific Objectives
Figure 1.17. Left: IXO 150 ksec observation of a low mass (kT=2 keV) cluster at z=1 with a bolometric
luminosity of 7.7 X 1043 erg\s, for two spectral extraction regions. In the 0.4R500-0.6R500 annulus (R500 is a
fiducial radius of the cluster within which the mean cluster mass density is a factor of 500 above the cosmic
critical density), the temperature and iron abundances are measured with an accuracy of ±5% and ±20%
respectively, illustrating the capability of IXO to measure temperature (and thus entropy) and abundance
profiles at z=1, even for low mass systems. Right: The same cluster but at z=2 observed for 250 ksec. Overall
temperature and abundances can be measured accurately: ±3% for kT, ±3.5% for O and Mg, ±25% for Si and S,
and ±15% for Fe.
How and when was the excess energy in the intergalactic medium generated?
One of the most important revelations from recent X-ray observations, supported by optical and IR
studies, is that non-gravitational processes, particularly galaxy feedback from outflows created by
supernovae and super-massive black holes (SMBH), must play a fundamental role, both in the history
of all massive galaxies and in the evolution of groups and clusters as a whole. Galaxy feedback is
likely to provide the extra energy required to keep the gas in cluster cores from cooling all the way
down to molecular clouds, to account for the energy (i.e. entropy) excess observed in the gas of groups
and clusters, to regulate galaxy and star formation, and to produce the galaxy red sequence (see also
section 1.1.3 discussing the physics of AGN feedback mechanism).
It is now well established from XMM-Newton and Chandra observations of local clusters and groups
that their hot atmospheres have much more entropy than expected from gravitational heating alone
Pratt et al. 2010; Sun et al. 2009). Determining when and how this non-gravitational excess energy
was acquired is an essential goal of IXO. Galaxy feedback is a suspected source, but understanding
whether the energy was introduced early in the formation of the first halos (with further consequence
on galaxy formation history), or gradually over time by AGN feedback, SN driven galactic winds, or
an as-yet unknown physical process, is crucial to our understanding of how the Universe evolved.
The various feedback processes, as well as cooling, affect the intergalactic gas differently, both in
terms of the amount of energy modification and of the time-scale over which this occurs. Measuring
the evolution of the gas entropy and metallicity (a direct probe of SN feedback), from the epoch of
cluster formation, is the key information required to disentangle and understand the respective role for
each process. Since non-gravitational effects are most noticeable in groups and poor clusters, the
building blocks of today's massive clusters, these systems are of particular interest.
A major challenging goal of IXO is thus to study the properties of the first small clusters emerging at
z~2 and directly trace their thermodynamic and energetic evolution to the present epoch. Future widefield Sunyaev-Zel'dovich, X-ray (e.g. eRosita) and optical-IR surveys (e.g EUCLID) will discover
many thousands of clusters with z<2, but will provide only limited information on their individual
18
1. Scientific Objectives
properties, especially at high z. These surveys will provide excellent samples of clusters for follow-up
studies with IXO at the high sensitivity and resolution required to determine the X-ray properties of
these low mass systems. In addition, 4 low mass clusters per deg2 with M>1013MO, will be detected
serendipitously within the 18’ X 18’ field of the IXO Wide Field Imager. The power of IXO is
illustrated in Fig. 1.17 that shows simulated, deep spectra for high redshift systems as will be obtained
with the calorimeter. These will provide gas density and temperature profiles, and thus entropy and
mass profiles to z ~1 for low mass clusters (kT~2 keV, Fig. 2, left) with a precision currently achieved
only for local systems. Measurements of the global thermal properties of the first poor clusters in the
essentially unexplored range z = 1.5-2 also will become possible (Fig. 1.16, right).
In conclusion, numerical simulations have reached a stage where modeling, including all hydrodynamical and galaxy formation feedback processes, is becoming feasible, although AGN feedback
modeling is still in its infancy. The appropriate physics of these processes is not well understood, and
advances are largely driven by observations. Thus, constant confrontation between numerical
simulations of galaxy cluster formation and observations is essential for making progress in the field.
IXO observations of the hot baryons, the most significant baryonic mass component of clusters,
combined with radio observations (e.g. SKA), observations of the cold baryons in galaxies (e.g. from
JWST, ALMA, and ELT) and of the dark matter via lensing data (LSST, EUCLID) will provide, for
the first time, the details for a sufficiently critical comparison. We expect that the major breakthrough
of a detailed understanding of structure formation and evolution on cluster scales will come from
simulation-assisted interpretation and modeling of these new generation observational data.
1.2.3 Galaxy cluster cosmology
Galaxy clusters as the largest well defined objects in the Universe form an integral part of the cosmic
large-scale structure (LSS). They are therefore very promising probes to assess the statistics of the
LSS, its growth over cosmic time, and lend themselves for tests of cosmological models. Systematics
in characterizing galaxy clusters by their mass is the most critical issue in using clusters for precision
cosmology. X-ray observations provide the most robust and detailed observables for cluster
characterization, and systematics can be further minimized by combination with observations at other
wavelengths. Historically galaxy clusters played an important role in establishing the current
cosmological paradigm in the early 1990's based on measurements of the cluster number density
(Henry & Arnaud 1991) providing evidence for a low amplitude of matter fluctuations and the baryon
mass fraction compared to Big Bang nucleosynthesis (White et al. 1993) pointing to a low matter
density. These results have subsequently be confirmed by the combination of other methods (Komatsu
et al 2010; Sanchez et al. 2009).
Two crucial observational approaches are used to test for the appropriate cosmological model and the
effect of Dark Matter and Dark Energy: (i) the measurement of the cosmic expansion history
(geometric method) applying e.g. to the observations of distant SN Ia and Baryonic Acoustic
Oscillations and (ii) the assessment of the growth of LSS via a time dependent growth factor used in
connection with observations of the galaxy distribution, cosmic gravitational lensing shear, and the
abundance and spatial distribution of galaxy clusters. The ESA/ESO Working Group Report on
Fundamental Cosmology highlights the complementarity of these methods and points out the
important role of galaxy cluster cosmology (once systematic uncertainties can be overcome). Both
types of cosmological tests are required to distinguish without major degeneracies between
quintessence types models and modified General Relativity for the explanation of the accelerated
cosmic expansion (Frieman et al. 2008; Sapone & Amendola 2007). Galaxy clusters are particularly
sensitive probes of structure growth through observations of their abundance (for given mass limit) as
function of redshift (Jenkins et al. 2001; Tinker et al. 2008). Clusters allow us to probe the redshift
19
1. Scientific Objectives
range z = 0 – 2 where the effect of Dark Energy is most significant. Therefore galaxy clusters studies
will form an indispensable part of the future precision cosmology efforts.
A number of galaxy cluster surveys ongoing and in preparation at different wavelengths (SZE
(Sunyaev-Zeldovich Effect): e.g. SPT, ACT and PLANCK; Optical/NIR: e.g. DES, PanSTARRS,
EUCLID and LSST; and X-rays: eROSITA) will produce large catalogs of galaxy clusters stretching
out to redshifts of 2, with very sparse characterization of the more distant clusters. Next generation
SZE experiments and the NIR sky survey by EUCLID will be most important to detect clusters at the
highest z. X-ray observations will still provide the essential information on the structure and mass of
clusters from detailed images and properties of the most massive baryonic cluster component (~85%),
the intracluster medium (ICM). A high throughput telescope with the specifications of IXO is required
to obtain precision data on cluster at redshifts beyond z=1.2. It will be important in two ways: (i)
improving the cosmological constraints of the mentioned surveys by significantly better calibrating
their cluster mass modeling and (ii) by using IXO follow-up to obtain precision X-ray parameters
such as ICM temperature, gas mass, and YX (= gas mass times temperature) for well selected cluster
samples.
Assessment of structure growth
The mass function of galaxy clusters, n(M), is an exponentially sensitive probe of the linear density
perturbation amplitude in the Dark Matter distribution and can be predicted with high precision by
theory for different cosmological models (Frieman et al. 2008; Sapone & Amendola 2007). Given
precise cluster masses, the perturbation growth factor can e.g. be recovered to 1% accuracy from a
sample of 100 massive clusters for given redshift (Vikhlinin et al. 2010). Given n(M) with an accuracy
of 1-2% (at z~1), the Dark Energy equation of state parameter, w, can e.g. be determined with 3-6%
uncertainty. Therefore, observations of the cluster mass function evolution provide a highly
competitive and more advanced assessment of cosmic structure growth than the two other most
promising methods, weak lensing tomography (Amara & Refregier 2008; Kitching et al. 2008) and
redshift-space distortions in galaxy clustering analysis (Guzzo et al.2008). The most critical link
between theory and observational test is the relation of easily observable cluster parameters and the
cluster mass.
Figure 1.18. Cosmological parameter constraints to be obtained
from follow-up observations of 2000 galaxy clusters with IXO in
the redshift range up to z=2 (Vikhlinin et al. 2010).
Recent studies have shown that X-ray observations can provide such tight relations for the parameters
X-ray luminosity (central region excised), temperature, and YX with a scatter of 17%, 10-12%, and 810% , respectively ( Kravtsiv et al. 2006; Allen et al. 2008; Pratt et al. 2009; Vikhlinin et al. 2006;
Mantz et al. 2009). This has made it possible with CHANDRA and XMM-Newton follow-up studies
of ROSAT detected cluster samples to gain interesting constraints on w, to provide evidence that the
growth of cosmic structure has slowed down at z < 1 (Borgani et al. 2001; Schuecker et al. 2003;
Mantz et al. 2008; Vikhlinin et al. 2009) and to place first constraints on possible departures from
General Relativity (Rapetti et al. 2009). Further important results come from the mass determination
comparison with lensing data and detailed hydrodynamical simulations which both point towards a
20
1. Scientific Objectives
low bias of X-ray cluster masses of few – 20%. Improving the mass estimates is the key to progress.
For instance, while the eROSITA survey is expected to provide e.g. constraints on w (dw/dz) of the
order 10% (50%) with present calibration, the results can be improved to 2% (12%) if external mass
calibration at the 1% level is available (Haiman et al. 2005). Combining lensing and X-ray
observations is crucial since X-rays yield the most certain detections of real clusters and lensing
provides a statistically unbiased mass measurement (Corless & King 2009) with large (~30%)
uncertainty. IXO observations and EUCLID, with lensing data for many Thousands of galaxy clusters,
will provide a large synergy for precise cluster mass calibration. IXO mass calibration: an
uncertainty of 1-2% is expected be achievable in the normalization of the mass scaling relation with
IXO & EUCLID. IXO will be crucial to reduce the intrinsic scatter in the relation, and thus
systematics in masses estimated from mass proxies, by allowing us to include further structure
parameters in the scaling relations, like the X-ray spectral line broadening as a measure of the
dynamical distortion of the clusters which affects the mass measurement. IXO follow-up: A major
follow-up program (~ 5-10 Ms) to get good observational mass proxy parameters (like YX) for ~ 2000
galaxy clusters with IXO contributing to the high z part up to z=2 can provide the constraints shown in
Fig .1.18 (Vikhlinin et al. 2009). In both cases can IXO help to significantly tighten the cosmological
parameter constraints, even though IXO is not an observatory focused primarily on cosmology. IXOSurvey: IXO will open a window for the discovery of an even more distant cluster population (z ~ 2)
in serendipitous cluster searches in IXO WFI pointings.
Further contributions to cluster cosmology studies
Massive galaxy clusters are large enough to capture fair samples cosmic matter reflecting the cosmic
ratio of dark to baryonic matter, with most baryons in X-ray luminous gas. Treating the gas mass
fraction as invariant with redshift (with some corrections derived from simulations) and using the
d(z)3/2 dependence of its measurement one can obtain further constraints on the geometry of the
Universe which are competitive to those of SN Ia observations as shown in Rapetti et al. (2008). In a
similar way will the combination of X-ray and SZE measurements provide cosmic geometry tests
(Bonamente et al. 2006). All these refinements rest on the ability to obtained precise and detailed Xray observations of galaxy clusters in the extended redshift range from z = 1 to 2, which is only
possible with IXO as illustrated in Fig. 1.17 (Arnaud et al. 2009). At z=2 the typical most massive
objects have several 1013 solar masses as the example of Fig. 1.17. Current observations in the redshift
range 1 to 1.4 are limited to a few of the brightest galaxy clusters obtained with very deep
observations.
1.2.4 Chemical evolution along cosmic time
The most abundant heavy elements in the Universe are the so-called alpha elements (O, Ne, Mg, Si, S,
Ar, Ca) and Fe-group elements. These elements are mostly released by supernovae (SNe), and some
are even created during the explosions. The production of alpha-elements is dominated by core
collapse SNe, the explosions of massive stars, whereas Fe-group production is dominated by Type Ia
(thermonuclear) SNe, which are thought to be exploding CO white dwarfs. The progenitor systems of
Type Ia SNe are still poorly understood. It is thought that there are two channels (Mannuci et al.
2006), one prompt channel in which the delay time between stellar birth and explosion is of the order
5x108 yr and a “tardy” channel with delay times of up to 4x109 yr. The relative importance of these
channels is not well understood, nor is it clear whether the designation “channels” is justified, or
whether there is a continuum of delay times. The designation channels may be justified, if the prompt
channel corresponds to single degenerate systems (a white dwarf with a non-compact secondary star)
or a double degenerate system (two white dwarfs merging, see Greggio 2010, for a discussion). Note
that there are still major problems in explaining both the Type Ia SN rate and the relative ratio of Fegroup over alpha-elements (de Plaa et al. 2007, Maoz 2008). According to stellar population and
binary evolution calculations the Type Ia rate and contribution to the ICM should be much lower than
21
1. Scientific Objectives
observed. Two other elements of interest, besides the alpha and Fe-group elements, are carbon and
nitrogen. The relative enrichment contributions from different stellar populations for carbon and
nitrogen are not well understood, as both low mass stars (AGB winds) as massive stars (stellar winds
and SNe) contribute. [Leave out carbon as it will be hard to detect in clusters? **Vink note]
X-rays provide one of the best means of studying these elements, as they have their K-shell transitions
in the energy range 0.1-10 keV, in addition Fe has its L-shell emission in the 0.7-1.3 keV range. The
line formation properties of these lines are well understood and are similar for all these elements. This
makes X-ray spectroscopy a powerful tool to measure abundances in tenuous gasses, and to make
progress in this field. X-ray spectroscopy will enter a new era with IXO, which will bring the study of
elemental abundances to a new level, both for those objects where the elements have just been
released, supernova remnants (see Section 1.4.2), and the largest reservoirs of hot gas in the Universe,
clusters of galaxies.
Clusters of galaxies are the largest gravitationally bound structures in the Universe, with most
baryonic matter in the form of hot, X-ray emitting gas (Section 1.2.2). Surprisingly, the metallicity of
the gas is about 50% of that of the Galactic interstellar medium, whereas the gas mass to stellar mass
ratio is one to two orders of magnitude larger than that in galaxies. This suggests either that the star
formation rate in cluster galaxies was much higher than in galaxies like our own, or galaxies do not
retain most of the elements they produce due to galactic winds and/or AGN outflows. Galactic winds
may be in particular important for less massive galaxies, which have shallower gravitational
potentials. Indeed, the metallicity of galaxies scales with galactic mass (Gilmore 2001). Since no gas
escapes from a given volume that collapses into a massive cluster, clusters contain all elements ever
produced by their member galaxies, and most of these elements are in the ICM instead of locked up in
stars. With IXO we can study these abundances as a function of cosmic times, corresponding to the
redshifts at which these clusters were formed (z~2) till the present time. Note that there is an important
connection between chemical enrichment and the entropy excess in clusters, as the feedback
mechanism that (pre)heated the ICM are likely to be the same mechanisms that ejected stellar and
explosive nucleosynthesis products into the intergalactic medium and ICM.
To sum up, the main questions that IXO will address concerning chemical evolution are:
 When were the alpha and Fe-group elements and carbon, nitrogen formed?
 How and at which redshifts were these mainly dispersed into the ICM?
 Can we identify the main production mechanism? More specifically:
o Can we separate the contribution of core collapse and Type Ia SNe?
o Can we identify the contribution from different Type Ia channels?
o What is the origin of nitrogen, AGB-winds or massive stars?
o Is there a contribution from ongoing low-level star formation inside clusters?
IXO can address these questions by making abundance maps of unprecedented quality of z=0. This
step forward is the results of IXO’s high spectral resolution of the calorimeters combined with its high
throughput. This will make it for instance possible to determine temperatures from line ratios alone
(e.g. K-alpha over K-beta line ratios), instead of just from the continuum shape. This will have a direct
influence on the accuracy of the abundance determinations. A new window on contributions from
different SN types will come from routinely measuring abundances of trace elements, such as
chromium, manganese, and nickel (see Section 1.4.2). The high quality abundance maps, together with
temperature maps, will help to identify locations of active enrichment, such as AGN driven outflows
(Fig. 1.19), or locations of recent ram pressure stripping. [Reference to current results?**Vink note]
IXO will also further extend abundance studies to fainter, less massive clusters and groups. This may
reveal trends in abundances as a function of cluster mass, and determine whether early star formation
was dependent on the initial overdensities prior to cluster/group formation [and it will help to test
22
1. Scientific Objectives
whether groups and less massive clusters retained all metals produced in their galaxies]. Another
important contribution that IXO will make is the determination of abundances in the low surface
brightness outskirts of clusters. The abundances there will be closer to the pristine abundances of the
intergalactic gas as it was prior to cluster formation.
Figure 1.19. Left: X-ray emission from the Virgo elliptical galaxy M86 and its 380 kpc ram pressure stripped
tail dominates this mosaic of Chandra images. Sensitive, high resolution X-ray spectroscopy will measure the
metallicity along the tail and in the surrounding cluster gas. Right: Chandra, XMM, and IXO simulations show
the metal abundance and distribution in a merging cluster. In this example, a merging 7 keV cluster of L X = 2 ×
1044 erg s-1 at z = 0.05 is observed for 30 ksec by each observatory compared to the actual assumed metal
distribution (lower right panel). In this example, the calorimeter will image a 0.3×0.3 Mpc region. A similar
cluster at z=0.1 would produce a comparable map in ∼ 100 ksec and the IXO field of view would be 0.54×0.54
Mpc. Several exposures would be required to map the whole region.
An even more important contribution of IXO will be its ability to measure abundances of clusters as
function of redshift, up to redshifts as high as z~2, the epoch of cluster formation (Fig. XX). Tentative
results up to z~1 suggest that the Fe abundance is steadily decreasing with redshift (Balestra et al.
2007, Maughan et al. 2008). This could be the result of the ongoing contribution of long delay channel
Type Ia SNe. However, it could also mean that Fe and other elements were already inside the clusters,
but confined to the galaxies, and were slowly released into the ICM by ram pressure stripping
(Kapferer et al. 2007), or by AGN driven outflows from galaxies. In particular, the giant central
elliptical galaxies may have retained the metals produced by their SNe for a long time, due to their
deep gravitational potentials. It will, therefore, also be important to measure the abundances of alphaelements as function of redshift, as these will show the enrichment contribution of core collapse SNe.
Since most core collapse SNe exploded before cluster formation, alpha-elements will make the
distinction between continuing enrichment by Type Ia SNe, or release of previous enriched gas from
cluster galaxies.
To summarize, the ICM provides a complete record of past stellar nucleosynthesis output of cluster
member galaxies. IXO will be important to improve our understanding of the ICM chemical
enrichment by extending abundance studies to the faint outskirts of clusters, and by providing accurate
measurements of fainter, less massive clusters and groups of galaxies. A major step forward in our
understanding will be provided by IXO’s ability to measure the enrichment history of clusters as
function of redshift for both alpha elements and Fe group elements. This will complement high
redshift studies of star formation and SN rates as function of cosmic time by future facilities as JWST,
ALMA and the ELT.
23
1. Scientific Objectives
1.3 Matter under extreme conditions
1.3.1 Strong gravity (black hole physics)
General relativity (GR) has been resoundingly tested in the weak field limit (by high-precision
observations in the Solar System, and of binary pulsars), but we still lack quantitative probes of it in
the strong-field domain. X-ray observations of active galactic nuclei (AGN) and stellar mass compact
objects (Galactic black-holes, GBHs and neutron stars, NSs) offer a unique opportunity for this since
they probe accretion and ejection flows in the regions closest to the event horizons of black holes,
where strong gravity effects (i.e., gravitational redshift, light bending and frame dragging) are
expected to shape the gas dynamics.
The high-throughput X-ray telescope IXO is mandatory for these studies. IXO will be the first
telescope with sufficient effective area to track the orbits of test particles near the event horizon,
which it will do by utilizing the Fe-Ka fluorescence line. In addition, IXO’s focal plane
complementary instrumentation will be decisive in permitting spin measurements in several
bright GBHs and AGN using at least five independent techniques. This will be key to “calibrate”
the spin measurements with FeK lines for their application on larger samples of fainter sources,
thereby constrain galaxy merger and accretion histories – important aspects of the co-evolution of
black holes and host galaxies (see Section 1.1).
1.3.1.1 Gas dynamics under strong gravity effects
Accretion disks, relativistic jets and high velocity and massive winds are known to be forming in the
innermost regions of AGN and GBHs, but do we understand the flow patterns and energy generation
around black holes? How do accretion disks launch winds and jets? These questions are intertwined
with those related to GR and only a significant step forward in observational capability will be able to
disentangle these important physical effects.
To perform physical tests of accretion and ejection mechanisms requires the ability to obtain excellent
spectra on relevant orbital and even sub-orbital timescales. This, in turn, requires a combination of
large collecting area, high spectral and timing resolution as offered by IXO only. IXO will enable us to
study the physics that drives disk accretion, and connections between disks, winds, and jets.
Probing General Relativity in the strong field regime
In the luminous black hole systems, the accretion flow is most probably in the form of a thin disk of
gas orbiting around the black hole. To a very good approximation, each parcel of gas within the disk
follows a circular test-particle orbit (e.g. Armitage & Reynolds 2003). This geometrical and
dynamical simplicity makes accretion disks useful for probing the black hole potential and,
hence, the predictions of GR.
Pioneering results obtained by ASCA on MCG-6-30-15 (e.g. Tanaka et al. 1995), and current studies
by XMM-Newton and Suzaku clearly show the broadening and skewing of the disk reflection features
in both AGN and GBHs/NSs predicted by GR (due to a combination of the relativistic Doppler shift
and gravitational redshift; for a review, see Miller 2007). In a small number of AGN, current
observations already hint at the power of orbit-by-orbit traces using emission lines (Fig. 1.20). With its
superior throughput, IXO will enable the detection of iron line variability on sub-orbital
timescales (from 100s to 1000s seconds) in approximately 20-30 AGN. Any non-axisymmetry in
the iron line profile (e.g. associated with the expected turbulence in the disk) will lead to a
characteristic variability of the iron line, with “arcs” being traced out on the time-energy plane (Fig.
1.21). GR makes specific predictions for the form of these arcs, and the ensemble of arcs can be fitted
for the mass and spin of the black hole, and the inclination at which the accretion disk is being viewed.
24
1. Scientific Objectives
A second kind of emission line variability will occur due to the reverberation (or “light echo”) of Xray flares across the accretion disk (see Section 1.3.1.2). Reverberation observations offer
unambiguous proof of the origin of the X-ray lines as reflection features, allowing us to map the
geometry of the X-ray source and inner accretion flow, and are key to determine the masses of
supermassive black holes. The path and travel time of photons close to the black hole is also strongly
affected by space-time curvature and frame-dragging. In systems with very rapidly rotating black
holes, the region of the accretion disk capable of producing line emission can extend down almost to
the event horizon, so we can probe time-delays along photon paths that pass close to the horizon.
These photon paths create a low-energy, time-delayed “tail” in the GR reverberation transfer function.
The nature of this tail is insensitive to the location of the X-ray source but is highly sensitive to the
spacetime metric. Their characterization will thus provide another potential test of GR, this time based
on photon orbits rather than matter orbits (Cunningham & Bardeen, 1973). The reverberation of
individual flares (Fig. 1.21) will be accessible to IXO in the brightest few AGN. However,
reverberation will be statistically detected in many more AGN and GBHs/NSs via the use of Fourier
techniques aimed at detecting the lag between the driving continuum emission and the strongest
fluorescent emission lines.
Figure 1.20. Current observations of AGN have
just barely been able to search for emission line
variations on the orbital timescale at the ISCO
(DeMarco et al. 2009). This figure depicts Doppler
shifts in an iron emission line consistent with
Keplerian orbits at the ISCO in the Seyfert AGN
NGC 3516. The saw-tooth pattern is exactly that
expected for orbital motion in the innermost
relativistic regime around black holes. IXO will be
the first observatory with sufficient area to measure
these motions in dozens of AGN covering more than
2 decades in black hole mass. (Figure taken from
Iwasawa, Miniutti, & Fabian 2004)
Figure 1.21. How does matter behave close to a
black hole? IXO will resolve multiple hot spots in
energy and time as they orbit the SMBH, each of
which traces the Kerr metric at a particular radius.
In the time-energy plane, the emission from these
hot spots appears as “arcs,” each corresponding to
an orbit of a given bright region. Note the arcs in
the trailed profile that can be directly mapped to
test particle like orbital motion of hot spots in the
disk. Bottom right panel : Result of a simulated
IXO observation, assuming a 3x107 Msun black hole
and a 2-10keV flux characteristic of a bright AGN.
(Figure adapted from Armitage & Reynolds 2003).
A video of an MHD simulation of a turbulent disk
at 30 degrees inclination, and the timeresolved iron
lines that it would produce, can be seen at:
http://ixo.gsfc.nasa.gov/documents/resources/poster
s/aas2009/brenneman_plunge_i30.avi
1. Scientific Objectives
25
X-ray “tomography” of disk inner regions
Time-resolved X-ray spectroscopy of several bright obscured AGN has revealed some clear cases of
eclipses of the X-ray source which lasted a few hours. They are thought to be due to obscuring clouds
with column densities of 1023-1024 cm-2 crossing the line of sight with velocities in excess of 103 km/s.
This opens up the possibility for a new, unique experiment of “tomography” of the X-ray source:
while passing, the cloud covers different parts of the X-ray source, revealing its structure (geometry,
emissivity, and associated relativistic effects). If an occultation by a Compton-thick cloud is observed
in a source with a broad relativistic line, the line profile is expected to change during the eclipse, as we
would probe the approaching and receding parts of the line emitting region separately (see Fig. 1.22).
Such experiment is just at reach of XMM-Newton or Suzaku for at least one object (NGC 1365), if
observed on a fortunate event. Conservative estimates predict that these measurements would be
relatively straightforward with IXO for several sources with great accuracy (see Fig. 1.22). This
experiment thus provides a unique opportunity to probe the relativistic effects due to strong
gravity and fast orbital motion in the innermost regions of AGN accretion discs using the X-ray
“tomography” of the disk line.
Figure 1.22. X-ray tomography of the emitting source and innermost regions of the accretion disk around the
central black hole in NGC1365 during eclipses. IXO simulations of the iron line profiles (shown as the ratio
between the data and the continuum best-fit model) at different phases of the disc occultation due to an orbiting
cloud. It is assumed that NGC 1365 is observed with IXO for ~200 ks, and that during this time a total eclipse
occurs, with a duration of 60 ks, in agreement with what was actually observed with XMM and Suzaku
observations. The relativistic line model is that obtained from the best fit of an XMM observation (Risaliti et al.
09). The green profile is obtained from the non-eclipsed part of the observation, while the red and blue profiles
correspond to the first quarter and fourth quarter of the eclipse, respectively, i.e. to the "left" and "right" halves
of the X-ray source. IXO will allow a detailed time-resolved analysis of such eclipses, providing a direct proof of
the general relativistic effects in the vicinity of the central BH.
X-raying Ultra-Fast Outflows
AGN outflows can release a huge amount of mechanical energy from the innermost region nearest the
black hole into the surrounding host galaxy environment. Indeed AGN outflows may be the key to
understanding the feedback process which regulates the growth of the black hole to that of the
host galaxy and subsequently how the first galaxies and primordial black holes evolved.
XMM-Newton and Chandra observations of nearby AGN and quasars have shown extraordinary
details below 2 keV of the so-called “warm absorbers” (Crenshaw, Kraemer & George ’03). These
have been interpreted in terms of multi-temperatures, and stratified, photo-ionized outflowing gas.
Unexpectedly, absorption lines at energies between ~ 7-10 keV have also been detected in about a
third of all radio-quiet and radio-loud AGNs observed long enough. These lines have been commonly
1. Scientific Objectives
26
interpreted as due to resonant absorption from Fe XXV and/or Fe XXVI associated to a zone of
circumnuclear gas photo-ionized by the central X-ray source, with ionization parameter log2–5
(with in erg s−1 cm) and column density NH1022–1024 cm−2. These extreme values and their typical
velocities of 0.1c (with values up to 0.3c), indicate the existence of previously unknown ultra-fast
outflows (UFOs) in AGN (Pounds et al. 2003; Tombesi et al. 2010).
At present it is possible to detect the UFOs in relatively low z (i.e. <0.1) bright AGN (with the
exception of a few gravitationally lensed broad absorption line QSOs), but with the superior
throughput and resolution of IXO, it will be possible to extend the study of these outflows out to high
redshift, where the bulk of the galaxy evolution occurs (Fig. 1.23, left and Section XX).
Whether produced by accretion disk winds/ejecta (e.g. King & Pounds 03) or connected to the base of
a possible weak jet (e.g. the “aborted jet” model by Ghisellini et al. 2004) is currently under debate.
Regardless their origin, the mass outflow rates of these UFOs appear to be comparable to the accretion
rate and their kinetic energy represent a significant fraction of the bolometric luminosity (Reeves et al.
2004). Therefore, they are in principle able to bring outward a significant amount of mass and energy,
which can have an important influence on the surrounding environment. In fact, feedback from the
AGN is expected to have a significant role in the co-evolution of AGN and their host galaxies (see
Section 1.1.3). High energy resolution and high throughput above 2 keV, such as planned for IXO,
would allow to resolve the line profiles and follow the line variations in time, thereby constraining
definitively the outflow properties (ionization, density, velocity, turbulence, abundance) and geometry
(Fig. 1.23).
Figure. 1.23. Left: Flux limits between 2-10 keV for a 5 detection of narrow absorption lines (with EW=10
eV) with the XMS in the 3–11 keV band. Right: XMS simulated 6-10 keV spectrum of a highly ionized and
massive absorber (UFO) as those found in nearby RQ and RL AGNs.
1.3.1.2 Measuring BH spins with X-rays using five independent
techniques
Ever since the seminal work of Penrose (1969) and Blandford & Znajek (1977), it has been realized
that black hole spin may be an important energy source in astrophysics, potentially allowing the
extraction of energy and angular momentum to be extracted from black holes themselves! Possibly,
but not necessarily, spin may have impact on the apparent radio-loud radio-quiet ‘dichotomy’ in AGN,
in the powering of jets and/or winds in AGN and GBHs (Rees et al. 1982), in the formation of gammaray burst engines, in producing gravitational waves from merging BHs, and may tell us about the final
stage of stellar evolution (Supernovae) and on larger scales on galaxy merger and accretion histories
(see Section XX).
1. Scientific Objectives
27
Despite its importance, however, we are only now gaining our first tantalizing glimpses of black hole
spin in a few objects. Unlike measurements of black hole mass (accurately obtained through
observations of the orbits of other bodies around the black hole such as stars or binary companions),
spin measurements require us to examine observables that originate within a few gravitational
radii of the black hole, and spectral signatures needed to determine it are only found in X-rays.
As shown below, IXO is ideally suited to measuring spin in stellar mass and supermassive black holes
through a variety of independent techniques.
Time -averaged fitting of FeK and FeL disk lines
Irradiation of the accretion disk by hard X-rays produces emission lines and a characteristic disk
reflection spectrum (for a review, see Miller 2007). The most prominent line is typically Fe K, due to
the high abundance and fluorescence yield of iron, but also FeL and lines from lower-z elements are
expected below a few keV (Fig. 1.24). The disk reflection spectrum is typified by an apparent flux
excess peaking between 20-30 keV, which is actually due to Compton backscattering. Relativistic
Doppler shifts and gravitational redshifts endemic to the inner disk around black holes act to skew the
shape of disk lines and the reflection spectrum. The shifts grow more extreme with increasing black
hole spin, as the innermost stable circular orbit (ISCO) extends closer to the black hole (Fig. 1.25).
The clear imprints of special and general relativity on the line profile are thus used to measure
black hole spin. Because the line shifts scale with gravitational radii (GM/c2), the mass of a given
black hole and its distance are not needed to measure its spin using this technique.
Relativistic disk lines are common in Seyfert AGN (e.g. Nandra et al. 2007), in stellar-mass black
holes (Miller 2007), and even in the spectra of accreting neutron stars (Cackett et al. 2010). The
commonality of disk lines permits comparisons of the relativistic regime around compact objects
across a factor of 109 in mass.
The effects of spin on the disk reflection spectrum are not subtle, but the disk spectrum must be
decomposed from other complexity such as continuum curvature or the effects of photoionized
absorbers. For this reason, current studies (with XMM-Newton and Suzaku) have been limited to a
handful of objects. The unprecedented sensitivity and energy resolution of IXO will overcome these
uncertainties and allow us to measure the spin in several tens of AGN, GBHs and NSs, going beyond
the nearest Seyfert 1 galaxies and the brightest GBHs. Fig. 1.25 illustrates the relationship between
spin, the ISCO, and relativistic disk lines.
FIGURE 2.1 Cappi
Figure 1.24. Left: An ionized reflection spectrum from before (top) and after (bottom) the relativistic smearing
effects (for a=0.7) are incorporated. The relativistically broadened disk reflection FeL and FeK features are a
powerful probe of black hole spin. Right: Ratio of the spectrum of the NLSy1 1H0707-495 to a simple
phenomenological continuum model which shows the ionized FeL and FeK relativistic lines around 0.9 keV and
6.5-6.7 keV, respectively. This is comparable to calculations shown in the left panel.
1. Scientific Objectives
28
FIGURE 2.2 Cappi
Figure 1.25 Left: Black hole spin can be measured by determining the innermost stable circular orbit (ISCO) of
an accretion disk. All methods of determining spin that use the disk rely on this relation shown here. Right: the
line profile in red is expected around a zero-spin Schwarzschild black hole; the more skewed line in blue is
expected around a maximal-spin Kerr black hole.
Reverberation
As mentioned above, measuring the light travel time between flux variations in the hard X-ray
continuum and the lines that it excites in the accretion disk provides a model-independent way to
measure black hole spin. The time delay simply translates into distance for a given geometry. If a
black hole has a low spin parameter, iron emission lines in the 6.4-6.97 keV range should have a
characteristic lag of approximately 6 GM/c3; if the black hole is rapidly spinning, lags can be as short
as 1 GM/c3. A clear evidence for such a lag, a reverberation lag of 30s between the direct X-ray
continuum and the FeL emission accompanying the relativistic reflection, has only recently been
unambiguously found (Fabian et al. 2009) in the NLSy1 1H0707-495, thanks to a week-long
observation with XMM-Newton (Fig. 1.26, left). Close to spinning black holes, the path light takes
will be strongly impacted by spacetime curvature. When very close to the black hole, an otherwise
isotropic source of hard X-ray emission will have its flux bent downward onto the disk. An observable
consequence of these light-bending effects is a particular non-linear relationship between hard X-ray
emission and iron emission lines (Miniutti & Fabian 2004). At present, there is tantalizing evidence
for this effect in some Seyfert AGN (e.g. Ponti et al. 2006) and stellar-mass black holes (e.g. Rossi et
al. 2005). The IXO HTRS (for GBHs and NSs) and the XMS/WFI (for AGNs) have the time
resolution, broad energy range, energy resolution, and flux tolerance needed to make careful studies of
lags that can lead to spin measurements and clear detections of gravitational light bending. A study of
the lags that can be detected with the HTRS is shown in Figure 1.26. The extraordinary sensitivity of
this instrument will enable errors of less than 1 GM/c3 for fluxes of approximately 1 Crab.
Figure 1.26. Left: A negative lag, such as that found above frequencies of 6 10-4 Hz or timescales shorter than
30 min, indicates that the harder flux, which is dominated by the power-law continuum, changes before the
softer flux, which is dominated by reflection (particularly the iron L line). Error bars, 1sigma. Right: Lag
measurements in AGN and GBHs/NSs and related uncertainties.
1. Scientific Objectives
29
Accretion disk continuum fitting
Thermal continuum emission from the accretion disk may be used to measure the spin of stellar-mass
black holes. An accretion disk around a spinning black hole is expected to be hotter and more
luminous than a disk around a black hole with low spin, because the ISCO is deeper within the
gravitational potential (see Fig. 1.24). New spectral models have recently been developed that exploit
the corresponding changes in the shape of the continuum to measure spin. If the mass and distance to a
black hole are known, these models may be applied to spectra in order to measure the spin of a stellarmass black hole (see, e.g., McClintock et al. 2006). By virtue of its flux tolerance, the HTRS aboard
IXO is best suited to measuring black hole spins using the disk continuum. As shown in Fig. 1.27,
the simulated statistical uncertainty on the black hole spin measured through continuum fitting is about
an order of magnitude smaller than the one derived by fitting the relativistic iron line profile, but both
measurements will be available simultaneously with IXO.
Figure 1.27. A typical high state spectrum of a
GBH (e.g. GX339-4) as observed by the IXOHTRS. Both a disk component and a weak iron
line are present in the spectrum, together with a
power law. Simultaneous fitting of the disk
component and line should yield similar values
for the black hole spins. The statistical accuracy
on the spin measurement from the disk component
is about ten times smaller than for the relativistic
iron line.
Quasi periodic oscillations
The X-ray flux observed from accreting X-ray binaries is sometimes modulated at frequencies
commensurate with Keplerian frequencies, close to the compact object. The oscillations are not pure
but quasi-periodic, have a small width due to small variations in frequency and phase as expected for
gas orbiting in a real fluid disk with internal viscosity. X-ray quasi-periodic oscillations (QPOs)
probe the strongly curved spacetime around compact stars, constrain the mass, spin and radius
of compact objects, and hold potential to test some of the most fundamental predictions of GR,
such as the existence of an innermost stable circular orbit. Observed frequencies follow a 1/M
scaling, strongly suggesting a relativistic origin. Although we currently lack a unanimously accepted
model for high frequency QPOs, most models indeed associate the QPOs with general relativistic
frequencies. This is the case for models involving 3 to 2 parametric resonance between epicyclic
frequencies. Their application to the existing observations points to maximally spinning black holes
(Fig. 1.28).
The large collecting area of IXO, combined with the energy range and time resolution of the HTRS,
make IXO an apt next-generation X-ray timing mission. The improved sensitivity will open the way to
detect strong QPOs on timescales closer to the coherence time of the underlying oscillator, detect the
weakest features predicted in models, allowing us to remove the degeneracy in their identification.
Moreover IXO will enable to detect QPOs in weaker and more distant sources, such as the still
puzzling ultra-luminous X-ray sources (possibly harboring intermediate mass black holes), and AGN,
for which a claim for a QPO detection was recently reported (Gierliński et al. 2008). In the IXO
timeframe, theoretical understanding of accretion disk physics as well as global disk simulations will
advance (e.g. QPOs are now beginning to appear in time dependent simulations of MRI turbulence).
This will provide the necessary framework for exploiting the potential of QPOs, for probing GR,
compact object parameters and accretion disk physics.
1. Scientific Objectives
30
Figure 1.28. Left: The three general relativistic frequencies around a 8 solar mass black hole with a
dimensionless spin parameter of 0.8. There exist two radii in the disk where the ratio of the vertical and radial
epicyclic frequencies are equal to 3:2 and 2:1. In a particular class of models, resonance should be excited at
these particular radii. Right) The HTRS power density spectrum simulated for an HTRS observation of 5
kseconds (0.8 Crab source), in which the 4 QPOs are simultaneously detected at a significance level larger than
6 sigmas. As of today, only the QPOs corresponding to the 3:2 ratio have been reported from RXTE
observations. If the mass of the black hole is known (from the dynamical mass function estimate of the binary
system), solving the GR frequency equations for the black hole spin should yield an estimate with an error of
about 10%.
Polarimetry
The polarization properties are significantly altered when photons travel on null geodesics in a strong
gravity field, as it happens in the vicinity of a black hole or a neutron star (Connors & Stark 1977). In
practice, this results in a rotation of the polarization angle as seen by a distant, and indeed it is also
known as “Gravitational Faraday Rotation” (Ishihara et al. 1998). The amount of rotation depends
on the geodesics parameters, and increases with decreasing impact parameter with respect to the
compact object. This effect can be used to measure the spin of the black hole in both GBHs and
AGN. In GBHs, when in soft state, emission is thermal from the accretion disk. Because the disk
temperature decreases with the disc radius, the higher energy photons are those mostly affected. As a
result, a variation of the polarization angle (Fig. 2.6, left), as well as of the polarization degree (Fig.
1.29, right) with energy is expected (Schnittman & Krolik, 2009, and ref. therein). This rotation
depends on the spin via the spin-dependence of the ISCO. In AGN, the disc thermal emission is much
softer, outside the X-ray band. However, strong gravity effects may manifest themselves through timedependent, rather than energy-dependent, rotation of the polarization angle of the Compton reflection
component (Dovciak et al 2004), which again is spin-dependent. In the light-bending model (Miniutti
& Fabian 2004), a relation with the flux of the source is also expected (Dovciak et al., 2004).
Figure 1.29. Energy-resolved polarimetry allows to estimate the spin of galactic black holes and provides an
independent tool to be used in complement to spectroscopy and timing (Dovciak et al., 2008)
1. Scientific Objectives
31
1.3.1.3 Sgr A* in the Galactic Center (**section to be moved)
Supermassive black holes are supposed to be present at the center of most (if not all) galaxies. At the
dynamical center of our Galaxy we find the closest supermassive black hole, Sagittarius A* (Sgr A*).
Its accretion luminosity is by several orders of magnitude lower than its Eddington luminosity. The
study of Sgr A* is important to investigate the evolution of accretion/ejection mechanisms around
supermassive black holes in general because they spend most of their time in a quiescent state. X-ray
flares from Sgr A* – with durations of up to a few hours – were brought to light by Chandra and
XMM-Newton (Baganoff et al. 2001, Porquet et al. 2003) and are believed to originate within just a
few Schwarzschild radii from the black hole event horizon. Thanks to the next generation instruments
aboard IXO, the accretion and/or ejection mechanisms at work in the close vicinity of this extremely
faint supermassive black hole can be explored in unprecedented detail. The WFI and the HXI will
allow to constrain the spectral properties of moderate to high amplitude X-ray flares from Sgr A*, and
to investigate any spectral variations during the course of the brightest flares. The X-ray polarimeter
will help to assess the geometry of the matter accreted and/or ejected. Finally, the large field-of-view
of the WFI and the HXI will enable us to observe simultaneously the neighborhood of Sgr A*, which
is one of the richest regions of the sky harboring numerous types of extended (e.g., diffuse emission,
supernovae remnants) and compact (e.g., X-ray binaries) astrophysical objects (see Fig. 1.30).
Figure 1.30. Simulation of a 28,000 s-observation of the Galactic
center centered on Sgr A* with IXO/WFI. The red, green, and
blue colors code the 1--3, 3--5, 5--8 energy bands, respectively.
The circle indicates the field-of-view of HXI. This simulation was
computed with the IXO simulator simx (version 1.0.0) from the
1,000,000 s exposure on Sgr A* obtained with Chandra (Muno et
al. 2009). [Courtesy of Nicolas Grosso]
Some possible additional text:
A case study to tackle the above issues is SgrA*, the supermassive black hole at the center of our
Galaxy, which is among the best examples of such starving black hole. Despite an estimated mass of
about 4e6 Msol, its total luminosity is only 300 times that of the sun, thereby implying both a very low
accretion rate and a radiative inefficient accretion flow. Despite being typically in a quiescent state,
SgrA* often displays about 1 flare per day in both X-rays and in the NIR bands. These flares are
intense, reaching tens to hundreds time their quiescent flux on short (hundreds to thousands seconds)
time scale. They are believed to originate very close to the SMBH, within few Rs from the horizon
(Baganoff et al. 2001, Genzel et al. 2003). The reason for these flares is currently not understood.
Most current hypotheses suggest either the sudden increase of the accretion rate, for example due to
blobs of material, or the sudden release of magnetic energy and electron accelerations in the flow, or
both. Current Chandra and XMM results indicate a spectral flattening during the flares, and that the
quiescent state is produced by a high temperature non-equilibrium plasma. Clearly, a high throughput
X-ray mission, with good enough angular resolution to isolate SgrA* from nearby Galactic sources,
will allow to follow the source spectral evolution during the flares, as well as its spectral and spatial
characteristics during the quiescent state. Such X-ray studies, in combination with simultaneous
observations at other wavelength, shall allow unprecedented constraints on models for flare production
and emission, as well as for radiatively inefficient flows.
Finally, SgrA* is also a case study for the study of SMBH activity in normal (i.e. non active) galaxies.
In fact, in the Galactic Center region, we can witness the recent past interaction of the SMBH activity
1. Scientific Objectives
32
with its surrounding environment. In particular there is evidence for SgrA* to have triggered and/or
interacted with the nearby supernovae remnant SgrA East. Another interesting interaction is with the
nearby giant molecular cloud Sgr B2 that may be reflecting in X-rays the past emission of the central
black hole, which should therefore have undergone a phase of activity about three hundreds years ago
(Koyama et al. 1996). If this is indeed the case, SgrA* was at that time a low luminosity AGN (Lx~
1039 erg/s), making it the brightest X-ray source in the sky after the Sun. Here an X-ray polarimeter
such as the one proposed for IXO will unambiguously tell. If the emission from the nebula is indeed
due to scattering, it should be very highly polarized (Churazov et al. 2002), with a direction of
polarization orthogonal to the scattering plane, and therefore to the line connecting SgrB2 and the
illuminating source, so unambiguously individuating the latter. The flux of SgrB2 is unfortunately
steadily decreasing over the years (Koyama et al.2008). Fortunately, there are other X-ray Reflection
Nebulae (XRN) in the Galactic Center also varying with time, so one expects that at the time a
polarimeter will be working, there will be a few bright enough for this test. In figure XX, an example
is also shown of the 6.4 keV line luminosity measured with Suzaku which suggests that XRNe are
indeed echoing the past activity of Sgr A* (Update with XMM results by Ponti et al. 2010). IXO will
allow to observe XRNe faint enough to be able to trace the activity of SgrA* over the past millenium,
thereby opening a unique window on the study of the activity of a SMBH and its interactions with its
hosting galaxy.
1.3.1.4 Cosmological Spins (**section to be moved)
After calibrating spin measurements with 5 different techniques, on a few tens of brigth GBHs and
AGNs, one can apply the FeK measurement of spin to a number of fainter sources, up to z about 1??
(TBD: Simulations of z=1 stacked spectrum with broad line component from COSMOS).
Spin measurements tell us about the final stage of stellar evolution (Supernovae) and on larger scales
on galaxy merger and accretion histories. In fact, the distribution of stellar-mass black hole spins
represents a rare and vital window on the central engines in SNe and GRBs and the first black holes in
the universe. Is it the degree of angular momentum imparted to the black hole that separates SNe and
GRBs? How much ionizing flux were the first black holes able to supply to young host galaxies?
Current X-ray observatories are beginning to measure spin parameters in a small number of sources.
The variety of techniques open to observers with IXO provides the best means of obtaining the
greatest possible number of spin measurements. Moreover, the sensitivity of IXO will make it possible
to obtain spin measurements in faint Galactic sources that are beyond the reach of current missions,
and make it possible to obtain spin measurements in bright stellar-mass black holes in nearby galaxies.
IXO will increase the number of current spin measurements by at least an order of magnitude,
revealing the nature of the central engine in GRBs and SNe, and the nature of the first black holes to
inhabit young galaxies at high redshift. Not clear about which caption goes
1. Scientific Objectives
33
Figure 4.1??. The histogram above shows a distribution of 10 black hole spins, measured using relativistic iron
emission lines from the accretion disk (Miller et al. 2009, Reis et al. 2009, Blum et al. 2009). IXO will increase
the number of stellar-mass black hole spin measurements by an order of magnitude, providing a unique window
on the central engine in SNe and GRBs. (Use latest similar compilation by Fender et al. 2010?? **note Cappi).
Figure 1.3.?? Spin as a probe of SMBH growth history. Left panels show the distribution of SMBH spin in a
scenario where seed black holes are formed at high redshift after which the SMBH population evolves purely via
SMBH-SMBH mergers. The middle panels show the spin distribution resulting from mergers plus standard
(disk) accretion. The right panels show the spin distribution when appreciable SMBH disk-accretion
accompanies merger events. The spin distribution is a powerful discriminant between growth histories that
may form identical mass-functions. With IXO we will measure the spin of few hundred SMBH and therefore
determine how they gained their large masses (Figure adapted from Berti & Volonteri et al. 2008).
1.3.2 Neutron star equation of state
The Fundamental Properties of Matter at High Densities
Seven decades after the first speculation on the existence of gravitationally bound neutron
configurations, we still know very little about the fundamental properties of neutron stars. Initial
attempts to model their mechanical properties were based on the assumption that the matter can be
adequately described as a degenerate gas of free neutrons, but it has become progressively clear that
the cores of neutron stars must in fact be the stage for intricate and complex collective behavior of the
constituent particles.
Over most of the range of the density/temperature phase plane, Quantum Chromodynamics (QCD) is
believed to correctly describe the fundamental behavior of matter, from the subnuclear scale up. The
ultimate constituents of matter are quarks, which are ordinarily bound in various combinations by an
interaction mediated by gluons to form composite particles. At very high energies, a phase transition
to a plasma of free quarks and gluons should occur, and various experiments are currently probing this
low-density, high temperature limit of QCD. Likewise, the QCD of bound states is beginning to be
quantitatively understood; recently, the first correct calculation of the mass of the proton was
announced.
The opposite limit of high densities and low (near zero, compared to the neutron Fermi energy)
temperature QCD has been predicted to exhibit very rich behavior. At densities exceeding a few times
the density in atomic nuclei ( ρ ~ 3  10 14 g cm 3 ), exotic excitations such as hyperons, or Bose
condensates of pions or kaons may appear. It has also been suggested that at very high densities a
phase transition to strange quark matter may occur. When and how such transitions occur is of course
determined by the correlations between the particles, and the ultra-high-density behavior of matter is
governed by many-body effects. This makes the direct calculation of the properties of matter under
these conditions from QCD extremely difficult. The only possible way to probe the high density, low
temperature limit of QCD is by observations and measurements of the densest material objects in
nature, neutron stars.
The Mass-Radius Relation of Neutron Stars
The relation between pressure and density, the equation of state, is the simplest way to parameterize
the bulk behavior of matter. It governs the mechanical equilibrium structure of bound stars, and,
conversely, measurements of quantities such as the mass and the radius, or the mass and the moment
of inertia of a star, probe the equation of state. Figure 1 shows the mass-radius plane for neutron stars,
with a number of possible mass-radius relations based on various assumptions concerning the equation
1. Scientific Objectives
34
of state (Lattimer and Prakash 2007). Two families of solutions have been indicated: the equations of
state for stars made up of bound quark states (baryons and mesons), and solutions for stars in which a
phase transition has converted most of the stellar matter to strange quark matter. The former stars are
gravitationally bound, and for a degenerate fermion gas, the radius generally decreases with increasing
mass of the configuration. The quark stars, on the other hand, are self-bound, and exhibit, very
roughly, an increase in radius with increasing mass. In the same figure, a number of constraints have
been indicated, derived from, for instance, the maximum stable mass of neutron stars, and a number of
broad constraints derived from observations (such as the high spin frequencies of two neutron stars). It
is obvious that definitive constraints can only be derived from simultaneous measurement of masses
and radii of individual neutron stars.
Effective discrimination between different families of hadronic equations of state will require a
relative precision of order 10% in mass and radius, and similar requirements apply to the strange
equations of state. In order to settle the question as to whether strange stars exist in nature, the
requirements depend on stellar mass. Since the hadronic and strange mass-radius relations cross in the
region ~ 1.3–1.8M , 12-16 km (ironically, those are the textbook values for the mass and the radius!),
we need to be able to probe a range of masses, or else have to rely on very difficult high-precision
measurements.
Figure 1.31. The mass-radius relationship for neutron stars reflects the equation of state for cold superdense
matter. Mass-radius trajectories for typical EOSs are shown as black curves. Green curves (SQM1, SQM3) are
self-bound quark stars. Orange lines are contours of radiation radius, R = R/ 1  2GM /Rc 2 . The dark blue
region is excluded by the GR constraint R > 2GM/c 2 , the light blue region is excluded by the finite pressure
constraint R > (9/4)GM/c 2 , and the light green region is excluded by causality, R > 2.9 GM/c 2 . The green
region in the right-hand corner shows the region R > Rmax excluded by the 716 Hz pulsar J1748-2446ad
(Hessels et al. 2006). From Lattimer and Prakash (2007).
Breakthrough Potential: X-ray Observations
Neutron stars have been the subject of intensive radio observations for forty years, and this work has
indeed produced a wealth of fundamental advances; probably the most famous among these is the
confirmation of the prediction of the gravitational wave power emitted by a relativistic binary based
on Einstein's quadrupole formula, which earned Hulse and Taylor a Nobel prize. As far as the
fundamental properties of the stars themselves are concerned, precise radio pulse arrival time
measurements on double neutron star binaries have produced a series of exquisite mass
1. Scientific Objectives
35
determinations, with a weighted average stellar mass of MNS = 1.413 ± 0.028 M (the error is the
weighted average deviation from the mean; see, for instance, Lattimer and Prakash 2007). But we need
the mass and the radius, or two other quantities derived from the internal structure, simultaneously.
There is currently no hope of measuring the stellar radii for the neutron stars for which we have a
precise mass, from radio or other observations (with one possible exception, see below).
Most of what we know about the fundamental properties of ordinary stars is based on a close study of
the emission spectrum emerging from their photospheres, and the same applies to neutron stars. The
natural wavelength band for photospheric observations is the X-ray band (neutron stars bright enough
for photospheric emission to be detectable are hot).
X-ray emission originating on the surfaces of neutron stars was first detected in X-ray bursts from
accreting neutron stars in Low-Mass X-ray Binaries (LMXBs), and photospheric emission has also
been detected from quiescent and isolated objects. These data had low spectral resolution and often
limited signal-to-noise; they have provided a very rough check on the order of magnitude of neutron
star radii, but precise measurement awaits the development of stellar atomic spectroscopy of neutron
stars, as well as a series of more exotic techniques that take advantage of general relativistic effects on
the surface emission. With observations performed with the diffraction grating spectrometers on the
Chandra and XMM-Newton observatories, this problem has come to the threshold of being resolved--the next step, based on sensitive, time resolved X-ray spectroscopy and energy-resolved fast
photometry has the unique potential of finally providing the window into QCD that the 'Cold Equation
of State' will open up.
The current observational situation is roughly the following. Evidence for atomic photospheric
absorption in the burst spectrum of at least one accreting neutron star (EXO0748  646), now appears
to have been spurious. The spin period of 45 Hz measured for this star has been redetermined to be
552 Hz, and rotational Doppler smearing should be severe enough to make photospheric absorption
lines currently not detectable. The distance to a number of hot, intermittently accreting neutron stars is
known, because they are located in Globular Clusters. Once accretion ceases, the atmospheres of these
stars should simply consist of pure H, and the spectrum can be calculated; this sample of stars with
known distance can be extended. Three apparently isolated neutron stars have a parallax measurement.
Several techniques are available with sensitive X-ray spectroscopy and fast photometry. Spectroscopic
observations of X-ray bursts give the atomic absorption spectrum, which, through pressure broadening
and GR effects, is sensitive to both the acceleration of gravity at the stellar surface, as well as the
redshift. Measuring two different functions of mass and radius thus gives mass and radius, separately.
For neutron stars at known distance, measured fluxes compared to the flux emerging from the stellar
photosphere will give the stellar radius. The continuum spectral shape is sensitive to the surface
gravity, again allowing a mass and radius measurement.
The ability to time- and energy-resolve emission from bursting or spinning stars provides unique
observational leverage. In particular, a capability to perform rapid (tens of microseconds or less)
spectroscopy allows resolving (and uniquely identifying) the severe Doppler broadening associated
with high stellar spin frequencies known to occur in most LMXBs.
1. Scientific Objectives
36
Figure 1.32. High resolution X-ray spectroscopy of the photospheric emission of a hot neutron star is
sensitive to the fundamental stellar parameters, through the effects of pressure broadening, relativistic
kinematics (rotation, Doppler shift, time dilation, beaming), and general relativity (light bending around the
star, gravitational redshift, frame dragging) on atomic absorption lines. The absorption line spectrum of a
1.4M neutron star, observed at 2 eV spectral resolution (left panel), in 120 sec of exposure of a moderately
bright X-ray burst with the microcalorimeter spectrometer as envisioned for the IXO mission. The spectrum
extends .... degrees in rotational phase. Emission is concentrated in a hot equatorial belt, seen at 5 degree
inclination. The absorption line is Fe XXVI H  . High time resolution spectroscopy can phase-resolve the
Doppler broadening of a rapidly spinning star (400 Hz) if the surface emission is azimuthally asymmetric. With
~ 100 eV energy resolution, and sub-msec time resolution (such as for the fast timing instrument on IXO), the
Doppler profiles in the right hand panel will be phase-resolved, allowing unambiguous determination of the line
broadening mechanism, and an absolute radius measurement (Fe XXVI Ly  ; same stellar parameters as
before).
For objects with a known spin period, the Doppler broadening provides a direct measurement of the
stellar radius. GR light bending effects on the phase modulation (of the absorption lines as well as the
total flux) again produce a measurement of the acceleration of gravity at the surface. In cases where an
atomic absorption line is detected, a redshift measurement is sufficient (with redundancy in the
pressure broadening). Note that the magnetic field strengths in LMXB's are small enough ( < 10 9 G)
that Zeeman splitting is not important.
High-speed, high-time resolution photometry of quasi-periodic intensity fluctuations associated with
the inner accretion disk in accreting neutron stars can yield absolute sizes, if light-travel time effects
can be resolved (reverberation). A time-delay spectrum within the frequency range of the variability
immediately provides the physical size of the inner rim of the accretion disk, where reprocessed soft
photons come from, and hence an upper limit to the neutron-star radius in km, independent of the
other stellar parameters. Likewise, Fe K line emission reverberation yields the same information.
There is multiple redundancy in several techniques, and we will have a choice of techniques to cover a
range of expected stellar spin periods (spin frequencies up to several hundred Hz have been measured
in LMXBs). Most of the necessary theoretical development (full radiative transfer neutron star
atmosphere models, effects of light bending on distant observer flux spectrum, etc.) is in place.
The precision required to make definitive measurements of the neutron star mass-radius relation is
within reach with currently feasible technology. The energy resolution of cryogenic X-ray
spectrometers, such as microcalorimeters, is sufficient to detect photospheric absorption lines and
measure their profiles. The count rate capability, time resolution, and CCD-style energy resolution of
Si drift detectors meet the requirements of fast energy-resolved timing. The International X-ray
1. Scientific Objectives
37
Observatory's capabilities, with a high energy resolution microcalorimeter spectrometer 2, and a high
time-resolution medium energy-resolution spectrometer3 are ideal for a definitive solution to the Cold
Equation of State problem. With the effective area, energy resolution, and timing capability of IXO,
the list of potential targets is at least a dozen deep for X-ray burst sources, and several quiescent
LMXBs will be observable as well.
Finally, the X-ray techniques will complement possible results from radio pulsar observations. The
binary radio pulsar PSR J 0737  3039 is known to exhibit a pulse arrival time evolution of one of the
two members that may signal relativistic spin-orbit coupling of the binary. If that is indeed correct, the
moment of inertia of this neutron star may be measurable, in addition to its mass, and these two
quantities together constrain the equation of state. The mass of this neutron star is close to the average
mass of observed for radio pulsar neutron stars, and so this measurement may have limited leverage on
the problem of distinguishing between hadronic and strange equations of state. Neutron stars in masstransferring binaries will give us access to a wider range of neutron star masses (of order a solar mass
of material can be transferred over the lifetime of an LMXB), to address this fundamental issue.
1.4 Life cycles of matter and energy in the Universe
Understanding the origins and distribution of matter and energy in the Universe is one of the most
important quests of astrophysics. Understanding the formation and evolution of the elements in the
Universe across the time will answer questions about when and where the majority of metals were
formed, how they spread and why they appear today as they are. Indeed matter is under constant
reformation through on-going star formation and nucleo-synthesis with a crucial role played both at
stellar and galactic scale by accretion and mass loss processes that drive stellar formation and
evolution across the entire HR diagram.
1.4.1 STARS AND PLANETS (**review section title)
1.4.1.1 X-ray emission from Young Stellar Objects
Young Stellar Objects, more evolved than Class 0, i.e. the very early protostars, are strong X-ray
sources. Understanding their X-ray emission is essential for studying the physics of their circumstellar
environment, in fact X-rays heat and ionize circumstellar material. Disk heating is important because it
provokes disk photo-evaporation. X-ray disk ionization (likely dominating cosmic ray ionization by
several orders of magnitude, Glassgold et al. 2000) is important because it couples the gas to the
magnetic fields in a sheared keplerian velocity field via the magneto-rotational instability, inducing
MHD turbulence, affecting angular momentum transport, with key effects on protoplanetary disk and,
eventually, on planetary system formation.
Accretion Phenomena in YSO: Only recently it has been found that soft X-ray emission (E ≤ 1 keV)
in some YSOs is likely due to accretion shocks. In young accreting stars (CTTS), material falls onto
the central star with velocities vpre ~300−500 km/s, forming a shock near the stellar surface. This
shock heats the infalling material up to temperatures of a few MK, making it X-ray bright (Gullbring
1994). X-ray luminosities of plasma heated in the accretion shock of a CTTS are predicted to be high
even for moderate accretion rates (i.e., LX ~ 1030 erg/s for M = 10−10 M○/yr, Sacco et al 2008).
2the
3the
X-ray Microcalorimeter Spectrometer, or XMS instrument
High Time Resolution Spectrometer, or HTRS instrument
1. Scientific Objectives
38
Therefore the soft X-ray emission from CTTS can be a direct probe of the accretion process, if its
coronal origin can be excluded. A key diagnostic to distinguish between coronal and accretion shocks
X-rays are densities: this requires high resolution X-ray spectroscopy. The Chandra and XMMNewton grating derived densities (using the He-like line triplets) of the 7 brightest and nearest CTTS
(e.g. Kastner et al. 2002; Argiroffi et al. 2007). are orders of magnitude higher than that of coronal
plasma and are in agreement with model predictions for accretion shocks. Thanks to the large
effective area in the 0.3−1.0 keV band of the IXO X-ray Grating Spectrometer (XGS) we can I) obtain
good S/N high resolution spectra, in reasonable exposure times (≤ 100 ks), for CTTS up to 500
pc (reaching and comparing all the major nearby star forming regions (SFRs; i.e. Taurus,
Chamaeleon, Orion etc.) significantly enlarging our knowledge; ii) obtain good S/N high
resolution spectra for the brightest and nearest CTTS on time scales of just 10 ks opening the
possibility to explore the variability of the accretion process (on predicted time scales ≤ 1 d) as well as
rotational modulation due to accretion stream shadowing; iii) measure with good S/N the line fluxes
of the Lyman series of Ovii and Oviii to constrain simultaneously source optical depth (which is
predicted and has been observed, Argiroffi et al.2009) and absorption due to surrounding material
(Brickhouse et al. 2010); with the XGS high resolving power (λ/∆λFWHM=3000) we can iv) measure
the line Doppler shift with respect to the stellar radial velocity, which is expected for the X-ray bright
accreting material, but not for the stationary coronal plasma (post-shock velocities are vpre/4 ~75−125
km/s; line centroid accuracy ∆v is σ/N, where σ is the standard deviation of the line profile and N the
line total counts, hence ∆v ≤ 10 km/s for N ~ 100).
Probing the YSO circumstellar disks with the Fe Kα fluorescent line: Fe Kα emission has now been
observed from about a dozen of YSOs, (e.g. Tsujimoto et al. 2005). The presence of accretion disks
and geometrical arguments led to the perception that the Kα lines from YSOs are produced in the
circumstellar disk not in the photospheres of the central stars, as is the case for the Sun or in other
evolved stars during intense flares (e.g. Culhane et al. 1981). Given the distances of YSOs only
Chandra and XMM-Newton CCD spectra are available where the 6.4 keV feature is barely separated
from the coronal 6.7 keV line. In the small sample of Fe Kα detections surprising discoveries have
been made. Especially the rare cases with time-resolved spectroscopy showed that the fluorescence
due to the thermal photo-ionization scenario may not always apply: i) An exceptionally large
equivalent width of 1400 eV incompatible with this scenario was measured in the rise phase of a flare
on a YSO in Orion (Czesla & Schmitt 2007). This observation can be reconciled with a photoionization origin only if part of the continuum X-ray emission is occulted, e.g. in a flare partially
hidden by the star (Drake et al. 2008); ii) In the YSO Elias 29 in the ρ Oph SFR the line was observed
during the quiescent state and not during a flare, suggesting a production mechanism unrelated to the
central star X-ray luminosity. Giardino et al.(2007) proposed that the observed Fe Kα emission from
Elias 29 may originate in star-disk loops from collisional excitation by non-thermal electrons. In this
latter scenario one expects to see a rotational modulation of the Fe Kα strength. High quality timeresolved X-ray spectroscopy is needed for a better understanding of the connection between the central
continuum X-ray source and the emission line. Time delays measured with reverberation mapping can
give access to the geometry of the system (cf Feigelson et al. 2009). The IXO/WFI will allow timeresolved spectroscopy studies: in Elias 29 the 6.4 keV line can be detected in only 25 ksec at much
higher significance than in the existing 10 times longer XMM-Newton observation. With typical
rotation periods for YSOs of ≥ 1 d, their rotational cycle can be well sampled with time-resolved
IXO/WFI spectroscopy. Given the extent of WFI field-of-view, the Fe Kα line can be simultaneously
detected from a large sample of YSOs. Moreover the IXO XMS, with its ~200 times improved
sensitivity, will detect (or put strong constraints on) X-ray irradiation in hundreds of YSO
circumstellar disks in all nearby SFRs, at the same time the IXO HXI will establish the intensity of
deeply penetrating hard (10-30 keV) X-rays needed to evaluate turbulence and disk “dead” zone.
1. Scientific Objectives
39
1.4.1.2 Mass loss, rotation and magnetic field of massive stars
The X-ray emission from massive stars is intimately related to their energetic winds and the linedriven instability (Feldmeier 1995). Analyses of the very broad (~ 1000 km/s) X-ray emission lines
have revealed valuable information on the spatial distribution of the wind X-ray sources and their
associated temperatures (e.g., Waldron & Cassinelli 2007). Hence, X-ray studies have the potential to
provide a deeper understanding of stellar magnetism, rotation, and wind structure, i.e. to elucidate
physics of mass loss in the upper part of the HR diagram. A key problem is the discordance of
deduced mass-loss rates owing to the systematic biases of traditional observational diagnostics (e.g.
Hα and UV lines, radio free-free emission) that often yield disagreements up to a factor 10 or more
(e.g., Fullerton et al. 2006). X-ray emission line profiles offer the unique opportunity to help resolve
the uncertainties in current determinations of mass loss. Two solutions have emerged to explain the
discordance in mass loss determination: stellar wind clumping (e.g., Oskinova et al. 2007) or wind
illumination by XUV radiation (Waldron & Cassinelli 2010). However the generality of these results
cannot be substantiated since only the brighest stars can be observed at high spectral resolution with
Chandra or XMM-Newton, and some bias cannot be excluded. High S/N lines profiles gathered with
IXO over a large representative sample of OB stars, will yield the wind distribution of hot plasma, the
mass-loss rates, the degree of wind inhomogeneity, and even the geometrical shape of wind clumps
(i.e., bow-shoch structures, Cassinelli et al. 2008), and finally resolve the mass-loss rate discordance
problem. This is vital to better understand stellar evolution, the physics of the massive stars and to
quantify massive star feedback.
The importance of magnetic fields in massive stars has been highlighted in the last few years by direct
measurements of surface fields up to kG strength (e.g. Donati et al. 2001, Bouret et al. 2008), but the
possibility to measure magnetic field directly today exists only for a few closest objects. The new data
have stimulated advances in the theory of dynamo in massive stars (Spruit 2002), and the inclusion of
magnetic fields in stellar evolution models (e.g. Maeder et al. 2009). The recent emergence of X-ray
spectroscopy as a potential tool to constrain magnetic properties of massive stars (e.g. Gagne et al
2005) is of crucial importance for modeling all phases of evolution from star formation to its collapse.
The observed cyclic variability of UV line profiles points out to the connections between magnetism
and rotation. The changes in strength and shape of X-ray emission lines is, potentially, the best tool to
explore the effect of magnetic field on wind structure, but current X-ray variability studies of OB stars
are hampered by the low photon rate. With IXO XMS it will be routinely possible to study the
variability of the X-ray emission lines over the relevant time scales and to probe its connection to
stellar rotation and magnetism.
Copious amount of X-rays is produced by collision of winds in massive binary stars (e.g. Stevens et al.
1992). The head-on collision of the winds often leads to a much harder X-ray emission than in single
massive stars. As a result, interacting wind systems frequently feature a strong Fe K emission line.
Observing the orbital changes of this line at high spectral resolution will allows mapping in X-rays the
wind interaction region, similarly to Doppler tomography studies performed in the optical domain (e.g.
Thaller et al. 2001). This will provide unique information on the conditions at the shock between the
stellar winds and on the efficiency of the cooling mechanisms. Understanding colliding wind binaries
is pivotal in inferring the properties of the massive, mostly in binary systems, fast evolving stars in the
Universe. IXO will allow us to extend these X-ray measurements to extragalactic massive binary stars
featuring different metallicities and hence different stellar wind properties.
1.4.1.3 Probing non-standard dynamos and emission mechanisms in Brown
Dwarfs
XMM-Newton and Chandra have established young Brown Dwarfs (BDs) as ubiquitous X-ray
emitters. However, the X-ray luminosities of evolved BDs are yet unknown since only one L dwarf
1. Scientific Objectives
40
has been detected in X-rays to date (Audard et al. 2007). As a result, the nature, the efficiency, and the
evolution of the dynamo – of which the coronal emission is a proxy – at work in substellar objects are
poorly known. Investigation of the occurrence of other X-ray emission mechanisms known to be at
work in T Tauri stars (TTS), e.g. accretion shocks (cf. Sec.4.1.1), have been limited by low count
statistics. Only IXO can address the following open questions: i) In TTS the LX/Lbol ratio is an
indicator of the strength of magnetic activity and of the underlying dynamo action, it spans the range
10−3...−5 with some scatter beyond these limits. In the BD regime, the LX − Lbol relation is
dominated by upper limits, so current data do not allow to assess on firm ground the possible drop of
the X-ray efficiency with respect to the TTS. Among the known young BD population in SFRs only
those with low extinction can be detected using present-day instrumentation at its very limit (cf. the
850 ks long COUP, Preibisch et al. 2005). With IXO/WFI we will reach a limiting flux of ~ 5 · 10 −17
erg/cm2/s in 200 ksec, and detect BDs in Orion (450 pc) down to LX =1028 erg/s even if absorbed by
AV=10 mag and about twice deeper for AV=5 mag. With IXO/WFI a majority of the upper limits will
be turned into detections and the substellar LX/Lbol relation be better constrained. ii) Evidence for a
possible age decay of BD X-ray emission is based on less than a handful of upper limits for BDs with
ages > 100 Myr (Stelzer et al. 2006). In analogy to the observed decline of chromospheric Hα
emission, the decline of LX could be due to the decrease of effective temperature with time for
substellar objects and the related inefficiency of coupling between the increasingly neutral matter and
the magnetic field. From the existing observational material, LX ≤ 1025 erg/s must be expected for
evolved BDs. With IXO/WFI the detection of such objects is feasible in just ~ 20 ksec out to 20 pc.
According to the census of nearby stars (Reid et al. 2008) ~100 L dwarfs and many more late-M stars
and BDs will be accessible to IXO. Iii) Given the low count statistics the properties of the X-ray
emitting plasma in BDs are poorly known. In particular, very few X-ray spectra of BDs outside SFRs,
with ages ≥ 10 Myr, are available (Rutledge et al. 2000; Tsuboi et al. 2003; Stelzer 2004). These
objects have mostly been detected during flares where emitting plasma is subject to additional heating,
making a comparison to quiescent spectra of younger BDs impossible. Similarly, for the young BDs
the total number of counts collected in a XMM-Newton or Chandra observation is typically <
200−300 counts. The importance of spectral analysis is evident from the recent Chandra detection of a
young BD (Stelzer et al. 2010) that has shown unusually high X-ray luminosity (LX ~ 1029.7 erg/s)
combined with unexpectedly soft temperature (kT ~ 0.23 keV). These properties are remarkably
similar to those of TW Hya, the prototype of an accreting TTS, whose X-ray emission is dominated by
the contribution from accretion shocks with only a minor coronal component. IXO/WFI will allow us
to sample statistical numbers of BD spectra in different activity levels and evolutionary phases, and
will let us find out how frequent soft, accretion dominated, X-ray plasmas are in the substellar regime.
1.4.1.4 The complex X-ray emission of solar system bodies
X-ray studies of solar system bodies have entered maturity thanks to the high sensitivity of XMMNewton and the sharp imaging of Chandra. Some of the processes taking place in planetary
magnetospheres involve highly energetic plasmas and powerful magnetic fields: electromagnetic
observations provide unique data to explore particle populations, their acceleration mechanisms and
their response to solar activity, as well as the conditions of magnetic fields anchored on fast rotating
bodies. These are 'next door' examples of extreme and widespread scenarios in astrophysics.
On Jupiter at least three different processes are known to produce X-rays (Branduardi-Raymont et al.
2007). The brightest emission comes from the aurorae near the poles, but we observe X-rays also from
lower latitudes, i.e. disk emission. The aurorae and the disk have different X-ray spectra. At soft Xrays (< 2 keV) the auroral spectrum is dominated by line emission, produced by charge exchange in
the interactions of highly stripped, energetic ions (oxygen and other elements) and hydrogen
molecules in the planet’s upper atmosphere. It is not clear yet whether the ion species originate from
1. Scientific Objectives
41
the solar wind (carbon) penetrating the magnetosphere, or from the volcanoes of Jupiter’s satellite Io
(sulphur). Determining the ions species (an easy task with IXO spectral resolution) would solve one
puzzle and would give us clues to the mechanisms of ion transport and acceleration. The X-ray
spectrum of Jupiter’s disk, instead, resembles that of the Sun, with strong iron and magnesium
emission lines, implying an origin in the scattering and fluorescence of solar X-rays. Finally, above 2
keV the Jovian spectrum turns into a smooth continuum, produced by electron bremsstrahlung. The
electron component varied significantly over the 3.5 day long devoted XMM-Newton observation,
probably in direct response to solar activity. IXO could explore to much higher energies and in greater
detail the electron population, its spectrum and acceleration, and its relation to the ion population.
Saturn, on the other hand, has shown no evidence (yet) for auroral X-ray emission; a combination of
scattering and fluorescence of solar X-rays is thought to be responsible for its disk, polar cap and ring
emissions (Bhardwaj et al. 2005a: Bhardwaj et al. 2005b; Branduardi-Raymont et al. 2010).
Fluorescent scattering appears to take place in the upper atmospheres of Venus (Dennerl 2008) and
Mars; both also possess an X-ray emitting exosphere, with that of Mars extending out to a few planet’s
radii, and displaying an exceptional richness of charge exchange emission lines (Dennerl et al. 2006).
A similar origin is attributed to X-rays from comets, the heliosphere, and the geocorona. The IXO
exceptional sensitivity, spectral resolving power, and FOV size well matched to planetary targets and
adequate spatial resolution will make giant strides in the investigation of the solar system.
1.4.1.5 Atmospheric evaporation in extra-solar planets
The X-ray luminosities of solar-type stars drop roughly by a dex between 107 and 108 yr, by another
dex between 108 and 109 yr, and more rapidly above 109 yr (e.g Micela et al. 1990; Preibish &
Feigelson 2005). The intense X-ray emission during early epochs, and the associated extreme
ultraviolet emission which is more difficult to study in young stars, will dissociate and ionize
molecules in planetary thermo-spheres and exospheres so that light atoms escape into the
interplanetary medium (cf. Güdel 2007; Penz, Micela & Lammer 2008). Solar wind, and flare particles
may also erode the entire atmosphere if no magnetic field is present, while host star X-rays will always
be an erosion source (Sanz-Forcada et al. 2010). These processes were probably important on Venus,
Earth and Mars during the first 108 yr and are presently leading to hydrodynamics escape of the
atmospheres in extra-solar “hot Jupiters.” The atmospheric conditions, and eventually the habitability,
of planets may thus be regulated in part by the evolution of the ultraviolet and X-ray emission of their
host stars. Thousands of extra-solar planets will be known by 2020 through Kepler mission and other
planetary search programs. IXO can measure both the quiescent and flare activity of specific stars
which will be known to have planets in their habitable zones. Combined with stellar activity
evolutionary trends and planetary atmospheric modeling, IXO findings should give unique insights
into the atmospheric history of these potentially habitable planets.
1.4.2 Supernova remnants: formation of the elements, shock heating and
particle acceleration
The most abundant massive elements in the Universe are released by, and in many cases formed in,
supernova (SN) explosions. These explosions are also the most important source of energy for the
interstellar medium, both in the form of kinetic/thermal energy and in the form of cosmic rays. IXO
will allow us to study these explosions, their products, as well as their immediate surroundings either
by directly observing X-ray emission from (extra-galactic) SNe, or by studying supernova remnants
(SNRs) in the Galaxy and Local Group. IXO’s non-dispersive high resolution imaging spectrometer
will allow us to map the temperatures and three dimensional distribution of alpha and Fe-group
elements in SNRs (see Sec XX), using Doppler shifts as the third dimension. A detailed knowledge of
SNe/SNRs is not only of interest as a topic in itself, but there are many synergies with the rest of the
IXO science case, as SNe are of key importance for the chemical evolution of the Universe, and often
1. Scientific Objectives
42
mark the creation of a neutron star or black hole. Moreover, the details of shock physics and cosmic
acceleration, which can be studied by IXO in nearby SNRs are of direct relevance to the large-scale
shocks that formed the WHIM and heated clusters of galaxies. Key questions that IXO will address
regarding SNe and SNRs are:
 What is the chemical production of SNe of different types, and what is its intrinsic variation?
 What is the origin of thermonuclear SNe (Type Ia SNe)?
 How do core collapse SNe (Type Ib/Ic/II/IIb) explode?
 What is the physics of collisionless shocks, and how do they accelerate particles?
Supernovae, explosion mechanisms, and nucleosynthesis products
Despite their importance for astrophysics we still lack fundamental knowledge about SNe. For Type Ia
SNe we still do not know which type of binaries containing a WD makes them, we do not what the
fundamental difference is between the long and short duration channels (Sec XX), and what
determines the observed variation in brightness. Since IXO will observe both the shocked
circumstellar medium and SN ejecta, IXO will probe both the SN explosion material and the
surroundings as shaped by the progenitor system. A crucial element in Type Ia SNRs is iron. Unlike
current instruments, IXO will be able to resolve the important Fe-L emission line complex around 1
keV. This will provide powerful plasma diagnostics, and will lead to accurate measurements of the
temperature, velocity (through Doppler shifts/broadening) and abundance structure. The modeling of
the SNR spectra have now advanced far enough to distinguish between energetic, normal and subenergetic Type Ia explosions, based on the iron content and energy content of the SNRs (Badenes et
al. 2008a). However, current and future theoretical models stress the importance of the 3D structure of
Type Ia SNe (Kasen et al. 2009), which only IXO can obtain using Doppler mapping.
For core collapse SNe the explosion process itself is not understood. Most of the energy released by
the collapse of the stellar core will be in the form of neutrinos. Part of the neutrino flux may drive the
explosion, but also rotation, magnetic fields, and acoustic instabilities may play a role in the explosion
process. This results in significant deviations from spherically symmetry. Young core collapse SNRs
have X-ray spectra dominated by O, Ne, Mg. The masses of these elements relate directly to the main
sequence mass of the progenitor. These O/Ne/Mg rich SNRs show pure metal ejecta of more massive
elements (Ar, Ca, Fe) that are distributed in an irregular pattern all over the SNR, and sometimes very
close to the shock front (e.g. Cas A and G292.0-1.8, see Hughes et al. 2000, Park et al 2007). This
suggests high velocities for complete and incomplete silicon burning products, which are synthesized
deeply inside the star, and whose velocities are not reproduced by current SN simulations (e.g.
Kifonidis et al. 2006). High spectral resolution imaging spectroscopy with IXO will make it possible
to reconstruct the 3D explosion properties by measuring Doppler shifts and broadening, even for
velocities as low as 300 km/s. Current X-ray detectors restrict Doppler measurements to velocities in
excess of 2000 km/s (for example Cas A, see Willingale et al. 2002, Delaney et al. 2010).
Recently, the study of nucleosynthesis has been extended to include trace elements as chromium and
manganese (Tamagawa et al. 2009, Fig. 1.33), extending our means to study the explosion physics. In
particular, the Cr/Mn ratio, but also the Fe/Ni, can be used to learn about the electron fraction during
nucleosynthesis. For Type Ia SNRs this is linked to the progenitor's original metallicity (Badenes et al.
2008b), whereas for core collapse SNe it tells us something about the exposure of the interior SN
ejecta to the extreme neutrino flux from the collapsing core. In addition, weak line emission can be
expected from radioactive 44Ti. This element is predominantly produced in core collapse SNe, when,
due to high expansion velocities deep inside the SN, the 4He fraction freezes out. This provides a
powerful diagnostic of the inner dynamics of the explosion, and the mass cut between SN ejecta and
neutron star. 44Ti has been detected in Cas A in gamma-rays (Iyudin et al. 1994) and hard X-rays
(Vink et al. 2001, Renaud et al. 2006), but it should also be visible in X-rays due to K-shell emission
from its daughter product 44Sc at 4.1 keV. The measured flux from Cas A translates into an IXO count
1. Scientific Objectives
43
rate of 0.25 cts/s. This is sufficient to map the spatial distribution and kinematics of 44Ti inside Cas A,
even for the unshocked, cold ejecta component. For SN 1987A, in the Large Magellanic Cloud, a
detectable flux is predicted of 0.04 ct/s, for the expected initial 44Ti mass of 10-4 Msun.
The high throughput of IXO will important for SNR population studies. Currently, our knowledge of
SNRs is based on the very biased sample of Galactic SNRs, and on the population of Magellanic
Cloud SNRs. IXO will be able to make a census of SNRs in the local group. Such a large sample will
give us a much better way of seeing the forest from the trees. Although detailed imaging may not be
possible, IXO’s spectral resolution will make it possible to accurate measure abundances, temperatures
and kinematic ages (from Doppler broadening). This will even allow for sub-typing of the SNRs. This
can, for example, be used to investigate the relation between Type Ia properties and local starforming
history (Badenes et al. 2009).
Figure 1.33. Left: Detail of the Suzaku spectrum of Tycho/SN1572, showing K emission from Cr and Mn
(Tamagawa et al. 2009). Right: Simulated IXO spectrum of the spatially integrated X-ray emission from a 400year-old remnant of dim (blue curve) and bright (red curve) SNIa in the Local Group galaxy M33 observed for
100 ks each.
Shock heating and particle acceleration
SNRs emit X-rays due to heating by high Mach number shocks that are driven into the surrounding
medium and back into the explosion debris. Like most astrophysical shocks these shocks are
collisionless, that is particle-particle collisions are rare and the heating occurs through plasma waves.
Apart from being a major heating source of the ISM, SNR shocks are probably responsible for most of
the cosmic rays observed on earth, at least for energies up to 1015eV. The theoretical expectation that
efficient cosmic ray acceleration modifies the heating and hydrodynamics of SNRs (Decourchelle et
al. 2000) has become clear over the last five years, mainly as a result of X-ray observations (e.g.
Warren et al. 2005, Helder et al 2009). Cosmic ray acceleration takes away energy, resulting in colder
plasma temperatures for a given shock velocity. Cosmic rays may also escape the SNR, thereby
extracting energy from the SNR shell. This results in a different hydrodynamic structure and evolution
of the SNR.
The temperatures of SNRs are in itself an interesting topic, as in most cases only the electron
temperatures are measured. However, collisionless shocks may heat ions to higher temperatures than
electrons. There is evidence based on optical (e.g. Ghavamian et al 2007) and X-ray spectroscopy
(Vink et al. 2003) that for shock velocities above 500 km/s electrons are indeed cooler than protons/
ions by factors of ten. In the optical these studies rely on the presence of neutral hydrogen atoms in the
circumstellar medium, whereas in X-rays the measurement could only be done for a particular bright
spot in SN 1006. Because IXO will offer imaging spectroscopy with high spectral resolution, thermal
Doppler broadening can be measured close to the edges of SNRs, thereby providing direct
measurements of ion temperatures. These can be compared to the more easily determined electron
temperatures. Moreover, combining these measurements with measured shock velocities will reveal
1. Scientific Objectives
44
whether shock energy has been transferred to cosmic rays, which will result in lower plasma
temperatures for a given shock velocity. This will be in particular interesting for those SNRs that show
X-ray synchrotron emission, a signature of fast cosmic rays acceleration. In fact, the X-ray emission
from some SNRs (RX J1713.7-3946 and “Vela Jr”) seems to be entirely caused by synchrotron
emission. Is this a result of electron temperatures below 106 keV, for which no X-ray thermal emission
is expected (Drury et al. 2009)? **note by Vink or is just the density in those SNRs very low, giving
rise to very weak line emission? The latter hypothesis can be tested by IXO, whose high throughput
and high spectral resolution will make it possible to detect weak line emission from regions dominated
by synchrotron emission. If detected, these lines can then be used to measure plasma temperatures and
densities.
Finally, IXO will improve our understanding of cosmic ray acceleration by identifying synchrotron
emission in hard X-rays. The hard X-ray synchrotron emission is expected to be very sensitive to
magnetic field fluctuations due to the steepness of the electron spectrum near the spectral cut off
(Bykov et al. 2009). Moreover, the emission from these flickering spots can be highly polarized. Hard
X-ray mapping and polarization measurements can therefore provide direct measurements of the long
wavelength modes magnetic field turbulence spectrum, which is an essential ingredient of acceleration
physics. IXO studies of the thermal and non-thermal emission from will be complementary to TeV
studies of SNRs with future gamma-ray telescopes such as the European Cherenkov Telescope Array
(CTA) and the US Advanced Camera Imaging System (AGIS). These observatories will show a
similar improvement in sensitivity over current observatories in TeV gamma-rays as IXO in X-rays.
The combination of X-ray and gamma-ray observations can be used to constrain the magnetic field
energy densities and maximum particle energies in SNRs.
1.4.3 Characterizing the ISM in the Galaxy
Except for the lightest species, the elements were produced through nucleosynthesis in stars. The
generations of stellar evolution responsible for the elements today are reflected in the relative
abundance of the elements, which is one reason that elemental abundance measurements is a central
issue in astronomy. Type II supernovae produce most of the α elements (e.g., oxygen) and some Fe,
Type Ia supernovae primarily produce iron-peak material, and mass loss from evolving stars
contributes additional C, N, and O. The abundances of these elements fix the relative number of the
different types of supernovae and the amount of star formation in a galaxy, and this has been a primary
tool for unraveling star formation histories in the Milky Way and other. If we are to comprehend the
evolution of the elements in the Milky Way, we must be able to determine elemental abundances in
both the stars and the gas, including the dust phase.
X-ray spectral observations provide a unique tool for the determination of elemental abundances.
Abundance determinations in the optical-UV rely on absorption or emission lines from particular ionic
states, with the subsequent large corrections for ionization state and depletion onto dust grains. In
contrast, X-ray observations can measure absorption across the ionization edge of an element through
inner-shell absorption, which is only weakly dependent on the ionization state of the gas. In addition,
dust is largely transparent to X-rays so the metals in grains are also revealed by their X-ray absorption
properties. Because of this combination of factors, abundances measured from X-rays are more
accurate and less subject to modeling uncertainties than abundances from optical-UV measurements.
These absorption techniques have been used for several sightlines, but a much richer future awaits the
sensitive observations possible with IXO.
High-quality spectra will not only determine elemental abundances, but ionization state distributions
and dust properties as well. For absorption at the ionization edge of a species, there is a shift in the
energy of the absorption edge with ionization state. The ionization “edge” really consists of a series of
edges, which if resolved, reveals the abundance of the individual ionic species along the line of sight.
1. Scientific Objectives
45
X-ray absorption studies can solve another long-standing problem: the actual composition of dust
grains. X-ray absorption features of various molecular solids are distinctive, so with good spectral
resolution, it will be possible to identify specific compounds, such as iron sulfate. It will become
possible to probe the composition of grains in various environments, providing critical constraints on
their formation.
Finally, X-ray observations will yield information on the sizes of dust grains, especially for the larger
grains, which account for most of the dust mass. This information is contained in X-ray scattered
light, which produces an X-ray halo around a bright point source. The halo can be measured as a
function of energy, leading not only to mass constraints on the large grains, but to their composition.
1. Scientific Objectives
46
REFERENCES
BARCONS (2)
Cosmic Vision: Space Science for Europe 2015-2025, ESA BR-247
De Zeeuw, P.T. & Molster, F., 2007, A Science Vision for European Astronomy, Astronet (www.astroneteu.org)
NANDRA (22)
Aird, J., Nandra, K., Georgakakis, A., Laird, E.S., Steidel, C.C., Sharon, C., 2008, MNRAS, 387, 883
Begelman M.C, Rossi E.M., Armitage P.J., 2008, MNRAS, 387, 1649
Begelman M.C., Volonteri M., Rees M.J., 2006, MNRAS, 370, 289
Bromm V., Loeb A., 2003, ApJ, 596, 34
Brusa, M., Comastri, A., Gilli, R., Hasinger, G., Iwasawa, K., Mainieri, V., Mignoli, M., Salvato, M., Zamorani,
G., Bongiorno, A., et al., 2009, ApJ, 693, 8
Croton, D.J., Springel, V., White, S.D.M., De Lucia, G., Frenk, C.S., Gao, L., Jenkins, A., Kauffmann, G.,
Navarro, J.F., Yoshida, N., 2006, MNRAS, 365, 11
Ebrero, J., Carrera, F. J., Page, M. J., Silverman, J. D., Barcons, X., Ceballos, M. T., Corral, A., Della Ceca, R.,
Watson, M. G., 2009, A&A, 493, 55
Fan, X., Narayanan, V.K., Lupton, R.H., Strauss, M.A., Knapp, G.R., Becker, R.H., White, R.L., Pentericci, L.,
Leggett, S.K., Haiman, Z., et al., 2001, AJ, 122, 2833
Fan, X., Strauss, M.A., Schneider, D.P., Becker, R.H., White, R.L., Haiman, Z., Gregg, M., Pentericci, L.,
Grebel, E.K., Narayanan, V.K., et al., 2003, AJ, 125, 1649
Ferrarese, L., Merritt, D., 2000, ApJ, 539, L9
Gebhardt, K., Bender, R., Bower, G., Dressler, A., Faber, S.M., Filippenko, A.V., Green, R., Grillmair, C., Ho,
L.C., Kormendy, J., et al., 2000, ApJ, 539, L13
Hopkins, P.F., Hernquist, L., Cox, T.J., Di Matteo, T., Robertson, B., Springel, V., 2006, ApJS, 163, 1
King, A.R., Pounds, K.A., 2003, MNRAS, 345, 657
Li, Y., Hernquist, L., Robertson, B., Cox, T.J., Hopkins, P.F., Springel, V., Gao, L., Di Matteo, T., Zentner,
A.R., Jenkins, A., et al., 2007, ApJ, 665, 187
Luo, B., Brandt, W. N., Xue, Y. Q., Brusa, M., Alexander, D. M., Bauer, F. E., Comastri, A., Koekemoer, A.,
Lehmer, B. D., Mainieri, V., Rafferty, D. A., Schneider, D. P., Silverman, J. D., Vignali, C., 2010, ApJS, 187,
560L
Marulli, F., Bonoli, S., Branchini, E., Moscardini, L., Springel, V., 2008, MNRAS, 385, 1846
Rhook, K.J., Haehnelt, M.G., 2008, MNRAS, 819
Salvaterra, R., Haardt, F., Volonteri, M., 2007, MNRAS, 374, 761
Silk, J., Rees, M.J., 1998, A&A, 331, L1
Silverman, J. D., Green, P. J., Barkhouse, W. A., Kim, D.-W., Kim, M., Wilkes, B. J., Cameron, R. A.,
Hasinger, G., Jannuzi, B. T., Smith, M. G., Smith, P. S., Tananbaum, H., 2008, ApJ, 679, 118
UEDA (16) Look up complete references
Hopkins et al. 2006
Worsley et al. 2005
Gilli et al. 2007
Maiolino et al. 2001
Imanishi et al. 2007
Risaliti et al. 1999
Ueda et al. 2007
Tueller et al. 2009
Daddi et al. 2007
Alexander et al. 2008
Fiore et al. 2008
Marconi et al. 2004
La Franca et al. 2005
Hasinger 2008
Fabian & Iwasawa 1999
Chartas et al. 2002
FABIAN (10)
Croton, D. et al. 2006, MNRAS, 365, 11
1. Scientific Objectives
47
Fabian, A.C. et al. 2003, MNRAS, 344, L433
Forman, W.R. et al. 2005, ApJ, 635, 894
McNamara, B.R. & Nulsen, P.E.J. 2007, ARAA, 45, 117
McNamara, B.R. et al. 2005, Nature, 433, 45
Peterson, J. & Fabian, A.C. 2006, Phys Rep, 427, 1
Reeves, J. et al. 2009, MNRAS, 2009, ApJ, 701, 493
Sanders, J. et al. 2008, MNRAS, 385, 1186
Silk, J. & Rees, M. 1998, A&A, 331, L1
Tremaine, S. et al. 2002, ApJ, 574, 740
BREGMAN (10)
Bregman, J.N. 2007, ARA&A, 45, 221.
Buote, David A., Zappacosta, Luca, Fang, Taotao, Humphrey, Philip J., Gastaldello, Fabio, Tagliaferri, Gianpier
2009, ApJ, 695, 1351.
Cen, R., & Fang, T. 2006, ApJ, 650, 573.
Cen, R., & Ostriker, J.P. 2006, ApJ, 650, 560.
Danforth, C.W., and Shull, J.M. 2008, ApJ, 679, 194.
Fukugita, M., & Peebles, P.J.E. 2004, ApJ, 616, 643.
Kaastra, J.S., Werner, N., Herder, J.W.A.d., Paerels, F.B.S., de Plaa, J., Rasmussen, A.P., & de Vries, C.P. 2006,
ApJ, 652, 189.
Nicastro, F., et al. 2005, ApJ, 629, 700.
Stocke et al., 2006, ApJ, 641, 217.
Werner, N., et al. 2008, A&A, 482, L29.
ARNAUD (4)
Markevitch, M. & Vikhlinin, A., 2007, PhR, 443,1
Clowe , D. et al., 2006, ApJ, 648, L109
Pratt, G., et al. 2010, A&A, 511, A85
Sun, M. et al. 2009, ApJ, 693, 1142
BOHRNINGER (28)
Henry, J.P. & Arnaud, K.A., 1991, ApJ, 372, 410
White, S.D.M., et al., 1993, Nature, 366, 429
Komatsu, E., et al. 2010, arXiv1001.4538
Sanchez, A., et al., 2009, MNRAS, 400, 1643NRAS, 400, 1643
ESO/ESA working group report on fundamental cosmology, ESO/ESA publication,
September 2006 (J. Peacock & P. Schneider, eds.)
Frieman, J.A., Turner, M.S & Huterer, D., 2008, ARA&A, 46, 385
Sapone,D. & Amendola, L., 2007, arXiv0709.2792
Jenkins, A., et al., 2001, MNRAS, 321, 372
Tinker, J., et al., 2008, ApJ, 688, 709
Vikhlinin, A., et al., Astro2010: The Astronomy and Astrophysics Decadal Survey, Science White Papers, no.
304
Amara, A. & Refregier, A., 2008, MNRAS, 391, 228
Kitching, T.P., Taylor, A. N. & Heavens, A.F. 2008, MNRAS, 389, 173
Guzzo, L., et al., 2008, Nature, 451, 541
Kravtsiv, A., Vikhlinin, A. & Nagai, D., 2006, ApJ, 650, 128
Allen, S.W., et al., 2008, MNRAS, 383, 879
Pratt, G., et al. 2009, A&A 498, 361
Vikhlinin, A., et al., 2006, ApJ, 640, 691
Mantz, A., et al., 2009, arXiv0909.3099
Borgani, S., et al., 2001, ApJ, 561, 13
Schuecker, P., et al., 2003, A&A, 402, 53
Mantz, A., et al., 2008, MNRAS,
Vikhlinin. A., et al., 2009, ApJ, 692, 1060
Rapetti, D., et al., 2009, MNRAS, 400, 699; 2010, MNRAS, 400, 699
Corless, V.L. & King, L.J., 2009, MNRAS, 396, 315
Haiman, Z., et al. 2005, astro-ph/0507013
Rapetti, D., et al., 2008, MNRAS, 388, 1265
1. Scientific Objectives
48
Bonamente, M., et al., 2006, ApJ, 647, 25
Arnaud, M., et al., 2009, arXiv0902.4890
VINK (8)
Balestra, I., et al., 2007, A&A 462, 429
De Plaa, J., et al., 2007, A&A 465, 345
Gilmore, G., 2001, in “Galaxies, Disks, and Disk Galaxies”, ASP Conf. Series 230, 3
Greggio, L., 2010, A&A, in press
Kapferer, W., et al., 2007, A&A 466, 813
Mannuci, F.,et al, 2006, MNRAS 370, 773
Maoz, D., 2008, MNRAS 384, 267
Maughan, B. J., et al., ApJS 174, 117
CAPPI (29)
Armitage, P., & Reynolds, C. S., 2003, MNRAS, 341, 1041
Blandford R. D., Znajek R. L., 1977, MNRAS, 179, 433
Cackett et al., 2010, arXiv:0908.1098
Connors P.A. & Stark R.F., 1977, Nature, 269, 128
Crenshaw, D.M., Karemer S.B. & George I.M, 2003, ARA&A, 41, 117
Cunningham J.M. & Bardeen C.T., 1973, ApJ, 183, 237
DeMarco B. et al., 2009, A&A, 507, 159
Dovciak M., Karas V., Matt G., 2004, MNRAS, 355, 1005
Dovciak M., et al., 2008, MNRAS, 391, 32
Fabian, A., et al., 2009, Nature, 459, 540
Ghisellini, G., Haardt, F. & Matt, G., 2004, A&A, 413, 535
Gierlinski M et al., 2008, Nature, 455, 369
Ishihara H., Takahashi M, Tomimatsu A., 1988, Ph. Rev. D., 38, 472
Iwasawa, K., Miniutti, G., & Fabian, A. C., 2004, MNRAS, 355, 1073
King A. R., & Pounds K.A., 2003, MNRAS, 345, 657
McClintock, J.E. et al., 2006, 652, 518
Miller, J. M., 2007, ARA&A, 45, 441
Miniutti, G., & Fabian, A., 2004, MNRAS, 349, 1435
Nandra K. et al. 2007, MNRAS, 382, 194
Penrose R, 1969, Riv. Del. Nuo. Cim, 1, 252
Ponti, G., et al., 2006, MNRAS, 368, 903
Pounds K.A. et al. 2003, MNRAS, 345, 705
Risaliti, G. et al. 2009, ApJ, 696, 260
Rees, M.J. et al., 1982, Nature, 295, 17
Reeves, J. N., O'Brien, P. T. & Ward, M. J., 2003, ApJL, 593, 75
Rossi, S. et al., 2005, MNRAS, 360, 763
Schnittman J.D., Krolik J.H., 2009, ApJ, 701, 1175
Tanaka, Y., et al., 1995, Nature, 375, 659
Tombesi F., et al., 2010, A&A, in press
CAPPI SgrA* ( blue marked by Cappi)
Baganoff et al. 2001, Nature, 413, 45
Muno et al. 2009, ApJS, 181, 110
Porquet et al. 2003, A\&A, 407, L17
Genzel, R., et al., 2003, Nature, 425, 934 Churazov E., Sunyaev R., Sazonov S., 2002, MNRAS, 330, 817
Koyama K., et al., 1996, PASJ, 48, 249
Koyama K., et al., 2008, PASJ, 60, 201
CAPPI Cosmological spin (TO BE DETERMINED)
PAERELS (2)
Hessels, J. W. T., et al., 2006, Science, 311, 1901
Lattimer, J. M., and Prakash, M., 2007, Physics Reports, 442, 109
SCIORTINO (43)
Argiroffi, C., Maggio, A., Peres, G. 2007, A&A, 465, 5
Argiroffi, C., Maggio, A., Peres, G., Drake, J. J., López-Santiago, J., Sciortino, S., Stelzer, B. 2009, A&A 507,
939
1. Scientific Objectives
49
Audard, M., Osten, R. A., Brown, A., Briggs, K. R., Güdel, M., Hodges-Kluck, E., Gizis, J. E.2007, A&A 471,
L63
Bhardwaj, A., Elsner, R. F., Waite, J. H.,Jr., Gladstone, G. R., Cravens,T. E., Ford, P. G. 2005a ApJ. 624, L121
Bhardwaj, A., Elsner, R. F., Waite, J. H.,Jr., Gladstone, G. R., Cravens,T. E., Ford, P. G.2005b ApJ. 627, L73
Bouret, J.-C.., Donati, J.-F., Martins, F., Escolano, C., Marcolino, W., Lanz, T., Howarth, I. D. 2008, MNRAS,
389, 75
Branduardi-Raymont, G., Bhardwaj, A., Elsner, R. F., Gladstone, G. R., Ramsay, G., Rodriguez, P., Soria, R.,
Waite, J. H., Jr., Cravens, T. E. 2007, A&A, 463, 761
Branduardi-Raymont, G., Bhardwaj, A., Elsner, R. F., Rodriguez, P.2010, A&A, 510, 73
Brickhouse, N. S., Cranmer, S. R., Dupree, A.K., Luna, G. J.M., Wolk, S. 2010, ApJ 710, 1835
Cassinelli, J., P., Ignace, R., Waldron, W. L., Cho, J., Murphy, N. A., Lazarian, A. 2008, ApJ, 683, 1052
Culhane, J.L., Rapley, C. G., Bentley, R.D., Gabriel, A. H., Phillips, K. J., Acton, L. W., Wolfson, C. J., Catura,
R. C., Jordan, C., Antonucci, E. 1981 ApJ 244, L141
Czesla, S., Schmitt, J. H. H. M 2007, A&A 470, L13
Dennerl, K. 2008, P&SS, 56, 1414
Dennerl,K., Lisse, C. M., Bhardwaj, A., Burwitz, V., Englhauser, J., Gunell, H., Holmström, M., Jansen, F.,
Kharchenko, V., Rodríguez-Pascual, P. M.2006, A&A, 451, 709
Donati, J., -F., Wade, G. A., Babel, J., Henrichs, H.F., de Jong, J. A., Harries, T. J. 2001, MNRAS, 326, 1265
Drake,J. J., Ercolano, B., Swartz, D. A. 2008, ApJ 678, 385.
Feldmeier, A. 1995, A&A, 299, 523
Feigelson,E., Drake,J., Elsner,R., Glassgold, A., Güdel, M., Montmerle, T., Ohashi, T., Smith, R.; Wargelin, B.,
Wolk,S. 2009, arXiv:0903.0598
Fullerton, A., Massa, D. L., Prinja, R. K. 2006, ApJ, 637, 1025
Gagne, M., Oksala, M. E., Cohen, D. H., Tonnesen, S. K., ud-Doula, A.; Owocki, S. P., Townsend, R. H.D.,
MacFarlane, J.. 2005, ApJ, 628, 986
Giardino, G., Favata, F., Pillitteri, I., Flaccomio, E., Micela, G., Sciortino, S. 2007, A&A 475, 891
Glassgold. A.E., Feigelson, E.D., Montmerle, T. 2000, in Protostar and Planets IV, 429
Güdel, M. 2007, Living Reviews Solar Phys., 4, 3
Gullbring, E. 1994, A&A 287, 131
Kastner, J. H., Huenemoerder, D. P., Schulz, N.,S., Canizares, C. R., Weintraub, D.A. 2002, ApJ 567, 434
Maeder, A., Meynet, G., Georgy,C., Ekström, S. 2009, IAUS 259, 311
Micela, G., Sciortino, S., Vaiana, G. S., Harnden, F. R., Jr., Rosner, R., Schmitt, J. H.M. M. 1990 ApJ, 348, 557
Penz, T., Micela, G. & Lammer, H. 2008, A&A 477, 309
Preibisch,T., McCaughrean, M. J., Grosso, N., Feigelson, E. D., Flaccomio, E., Getman, K., Hillenbrand, L. A.,
Meeus, G., Micela, G., Sciortino,S., Stelzer, B. 2005, ApJS 160, 582
Preibisch, T., Feigelson, E. D. 2005, ApJS 160, 390
Oskinova, L.,M., Hamann,W.-R., Feldmeier,A. 2007, A&A, 476, 1331
Reid, I.N., Cruz, K. L., Kirkpatrick, J. D., Allen, P. R., Mungall, F., Liebert, J., Lowrance, P., Sweet, A. 2008,
AJ 136, 1390
Rutledge, R. E., Basri, G., Martín, E. L., Bildsten, L. 2000, ApJ 538, L141
Sacco, G. G.; Argiroffi, C., Orlando, S., Maggio, A., Peres, G., Reale, F. 2008, A&A 491, 17
Sanz-Forcada, J., Ribas, I., Micela, G., Pollock, A. M. T., García-Álvarez, D., Solano, E., Eiroa, C. 2010 A&A,
511, L8
Spruit, H.C. 2002, A&A 381, 923
Stelzer, B.2004, ApJ 615, L153
Stelzer, B., Micela, G., Flaccomio, E., Neuhäuser, R., Jayawardhana, R. 2006, A&A 448, 293
Stelzer, B. et al. 2010, MNRAS, submitted.
Stevens, I.R., Blondin,J.M., Pollock, A. M. T. 1992, ApJ 386, 265
Thaller, M. L., Gies, D. R., Fullerton, A. W., Kaper, L., Wiemker,R.2001, ApJ 554, 1070
Tsuboi, Y., Maeda,Y., Feigelson, E. D., Garmire, G. P., Chartas, G., Mori, K., Pravdo, S. H. 2003, ApJ 587, L51
Tsujimoto, M., Feigelson, E.D., Grosso, N., Micela, G., Tsuboi, Y., Favata, F., Shang, H., Kastner, J. H., 2005,
ApJS 160, 503
Waldron, W., & Cassinelli, J. 2007, ApJ, 668, 456
Waldron, W., & Cassinelli, J. 2010, ApJ, 711, L30
VINK Supernova (19)
Badenes, C., et al., 2008a, ApJ 680, 1149
Badenes, C., et al., 2008b, ApJ 680, L33
1. Scientific Objectives
Badenes, C., et al., 2009, ApJ 700, 727
Bykov, A., et al., 2009, MNRAS 399, 1119
Decourchelle, A., Ellison, D.C., Ballet, J., 2000, ApJ 543, 57
DeLaney, T., et al., ApJ, in press
Drury, L. O’, et al., 2009, A&A 496, 1
Ghavamian, P., et al., ApJ 654, L69
Helder, E.A. et al., 2009, Science 325, 719
Hughes, J. P., et al., 2000, ApJ 528, L109
Iyudin, et al., 1994, A&A 284, L1
Kasen, D., Röpke, F. K., Woosley, S. E., 2009, Nature 460, 7257
Kifonidis, K., et al., 2006, A&A 453, 661
Park, S., et al., 2007, ApJ 670, L121
Renaud, M., et al., 2006, ApJ 647, L41
Tamagawa, T., et al., 2009, PASJ 61, S167
Vink, J., et al., 2003, ApJ 587, L31
Vink, J., et al., 2001, ApJ 560, L79
Warren, J., et al, 2005, ApJ 634, 376
BREGMAN “ISM” (none included)
50