Advances in Space Research 38 (2006) 1501–1508

www.elsevier.com/locate/asr

EUV spectroscopy in astrophysics: The role of compact objects

K.S. Wood a, M.P. Kowalski a,*
, R.G. Cruddace a, M.A. Barstow b

a
Naval Research Laboratory, Code 7655, 4555 Overlook Avenue, SW, Washington, DC 20375, USA
b
Department of Physics and Astronomy, University of Leicester, University Road, Leicester LE1 7RH, UK

Received 30 September 2004; received in revised form 3 May 2005; accepted 20 February 2006

Abstract

The bulk of radiation from million-degree plasmas is emitted at EUV wavelengths. Such plasmas are ubiquitous in astrophysics, and
examples include the atmospheres of white dwarfs, accretion phenomena in cataclysmic variables (CVs) and some active galactic nuclei
(AGN), the coronae of active stars, and the interstellar medium (ISM) of our own galaxy as well as of others. Internally, white dwarfs are
formally analogous to neutron stars, being stellar configurations where the thermal contribution to support is secondary. Both stellar
types have various intrinsic and environmental parameters. Comparison of such analogous systems using scaled parameters can be fruit-
ful. Source class characterization is mature enough that such analogies can be used to compare theoretical ideas across a wide dynamic
range in parameters, one example being theories of quasiperiodic oscillations. However, the white dwarf side of this program is limited by
the available photometry and spectroscopy at EUV wavelengths, where there exist critical spectral features that contain diagnostic infor-
mation often not available at other wavelengths. Moreover, interstellar absorption makes EUV observations challenging. Results from
an observation of the hot white dwarf G191-B2B are presented to demonstrate the promise of high-resolution EUV spectroscopy. Two
types of CVs, exemplified by AM Her and EX Hya, are used to illustrate blending of spectroscopy and timing measurements. Dynamical
timescales and envisioned performance parameters of next-generation EUV satellites (effective area >20 cm2, spectral resolution >10,000)
make possible a new level of source modeling. The importance of the EUV cannot be overlooked given that observations are continually
being pushed to cosmological distances, where the spectral energy distributions of X-ray bright AGNs, for example, will have their max-
ima redshifted into the EUV. Sometimes wrongly dismissed for limitations of small bandwidth or local view from optical depth limita-
tions, the EUV is instead a gold mine of information bearing upon key issues in compact objects, but it is information that must be won
through the triple combination of high-spectral resolution, large area, and application of advanced theory.
Ó 2006 COSPAR. Published by Elsevier Ltd. All rights reserved.

Keywords: Extreme ultraviolet; Astrophysical spectroscopy; White dwarfs

1. Introduction made the surprising discovery that the spectrum is that of

a white star not very different from Sirius itself. . . . we find
White dwarfs were the original compact objects. Armed . . . the radius is 18,800 km. Apparently, then we have a star
only with pre-Fermi physics Eddington (1926) described of mass about equal to the sun and of radius much less
the belated realization of the extremely small size of Sirius than Uranus. The calculated density is 61,000 g cm3, just
B, some 60 years after its discovery: ‘‘The mass . . . is found about a ton to the cubic inch. This argument has been known
from the double star orbit and is quite trustworthy. The for some years. . . I think it has generally been considered
absolute magnitude is 11.3, corresponding to a luminosity proper to add the conclusion ‘which is absurd.’. . . I do not
1/300 of that of the sun. The faintness would occasion no see how a star, which has once got into this compressed
surprise if this were a red star; but in 1914 W.S. Adams condition is ever going to get out of it. So far as we know,
the close packing of material is only possible so long as the
temperature is great enough to ionize the material. When
*
Corresponding author. the star cools down and regains the normal density ordi-
E-mail address: [email protected] (M.P. Kowalski). narily associated with solids it must expand and do work

0273-1177/$30 Ó 2006 COSPAR. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.asr.2006.02.043
1502 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508

against gravity. The star will need energy in order to cool. moment, and accretion environment. With few exceptions
. . .’’ (and those exceptions are instructive), to each white dwarf
From this start, white dwarfs have become keystones to observational class or phenomenon, there corresponds an
understanding stellar evolution in the Galaxy. Over 90% of analogous neutron star category or phenomenon. One
its stars evolve into white dwarfs. For temperatures of needs to scale the physics properly incorporating differenc-
80,000–300,000 K, white dwarf spectra peak in the FUV es in mass, radius and other parameters to recognize and
and EUV, and optical data provide comparatively less develop the correspondences. One instructive example is
information. Photometric and spectral observations with the correspondence between classical novae in white dwarfs
EUVE (Extreme Ultraviolet Explorer) have produced a and X-ray bursts in neutron stars. In each case, the com-
wealth of information about their atmospheres, but results pact object in a binary accretes material of roughly cosmic
were limited by its modest effective area and spectral abundances, which settles into a surface layer until pressure
resolution. at its base becomes critical and thermonuclear runaway
ensues. Historically, familiarity with classical novae
2. White dwarf – neutron star correspondences brought quick recognition of what was happening in (type
1) X-ray bursts. In general, theoretical understanding in
Nuclear burning has ceased in white dwarfs, except in either family should extrapolate properly to the other.
the atmosphere when the star is accreting. (Nevertheless Table 1 lists correspondences between the white dwarf
the nuclear burning that does occur in surface layers that and neutron star families.
make up only about 105 of the stellar mass accounts for Where symmetry of the correspondence is broken, the
classical novae, which in turn contribute much or most of white dwarf side often seems more interesting: white
the 26Al in the Galaxy!) White dwarfs have a stable config- dwarfs, because of different initial conditions, have a wider
uration and balance gravity with electron Fermi pressure, range of core compositions (He, CO, Ne, etc.) and hence a
so that there are 2 electrons per phase space volume h3. wider range of final masses. Neutron stars initialize at mass
By analogy, one could have a star with 2 neutrons per 1.4 Mx and usually remain near that value except when
h3, and this analogy was pursued theoretically (Oppenhei- subject to prolonged accretion; white dwarf initial masses
mer and Volkoff, 1939). Yet the discovery of neutron stars range over a factor of two. White dwarf spectral energy dis-
in the 1960s managed to be once again a surprise, because tributions peak where many atomic transitions also occur,
to skeptics that analogy had seemed untrustworthy. making their spectra more interesting and challenging espe-
The white dwarfs and the neutron stars are now the two cially in the EUV. White dwarfs can have larger magnetic
great families of Fermi-pressure-supported compact moments than neutron stars, e.g., AM Her is a stronger
objects. Because no suitable fermions exist beyond neu- ‘‘magnet’’ (in the sense of magnetic moment) than a magn-
trons, greater densities and relativistic degeneracy produce etar neutron star. Isolated white dwarfs may have ‘‘levitat-
black holes. Analogies and correspondences between the ed’’ atmospheres in which the effects of radiation pressure
families have proliferated in ways no one could have antic- can be pursued species by species. Their photospheres are
ipated. Both neutron stars and white dwarfs exist in many more accessible to observation. White dwarfs are numer-
combinations of parameters such as spin period, magnetic ous; there are plenty of nearby targets supporting appropri-

Table 1
Correspondence between white dwarfs and neutron stars
White dwarf family Neutron star family
Support by electron Fermi pressure, R  3000–12,000 km Support by neutron Fermi pressure: R  10 km
Magnetic, isolated white dwarf Crab pulsar
Rapidly rotating magnetic white dwarf Millisecond pulsars
PG1159 (not clear)
Non-pulsing isolated white dwarf, hot Non-pulsing central star in SNR
Non-pulsing isolated white dwarf, cool No analog, too faint to have been seen
Sirius B Non-accreting binary neutron star, e.g., Be star
U Gem, SS Cyg LMXB, atoll
(not clear) LMXB, Z, accreting near Eddington limit
Classical nova X-ray burst
Dwarf nova, Z Cam (various possibilities, including Rapid Burster)
Largest nova? X-ray superburst?
Doppler-broadened disk lines Doppler-broadened disk Fe lines –still uncertain
Short binary period (AM CVn): gravitational radiation Short binary period (X1820-303): gravitational radiation
CV orbital period anomalies in ‘‘O–C’’ ‘‘O–C’’ effects in EXO 0748-676
DQ Her (intermediate polar) HMXB, Her X-1, Cen X-3
AM Her (polar) No analog, would be magnetar in binary
Type Ia Supernova Neutron star – black hole by accretion from companion (rare)
 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508 1503

ately detailed observation, to be compared with similarly

detailed numerical modeling. As another illustration, white
dwarfs are thought to be progenitors of the cosmologically
important type Ia supernovae yet this widespread notion
has never been pursued to the point of successfully identi-
fying the progenitors with a specific known stellar popula-
tion. That unfinished business raises questions regarding
utilization of the SNe as cosmological probes.
So much observational and theoretical work has been
done on neutron stars that more open questions exist on
the white dwarf side than on the neutron star side. As will
be seen, these questions have gone unanswered largely for
lack of EUV data. It is not that white dwarfs can only be
seen in this channel; it is rather that many crucial observa-
tions have to be made there, and this in turn is primarily
because of the temperatures characteristic of young white
dwarfs.

3. Spectral energy distributions Fig. 1. Observed (points with error bars) and theoretical (continuous
curves) white dwarf luminosity functions. The theoretical curves are for
different white dwarf cooling models. Curves are calculated for different
Neutron stars emit by various mechanisms, one of which
values of disk age in 2-Gyr steps.
is thermal emission directly from the surface. The luminos-
ity grows with temperature as T4 until the Eddington limit
is reached, at which point the surface temperature is slight- burn were reached. This slow cooling evolves the spec-
ly over 2 keV. At this temperature, the associated radiation trum downward in frequency through a succession of
pressure will seriously retard any accretion, and if it some- wavebands, making white dwarfs the ideal star for the
how becomes hotter temporarily it can cause outflows. This working out of all stellar evolution in the Galaxy. Most
is also the temperature where the surface becomes unstable valuably, it represents fossilized information bearing
to radiation, leading to levitation seen in X-ray bursts as an upon the past history of production of compact objects.
expansion of the stellar radius. The analogous temperature The theoretical curves shown in Fig. 1 are only the ones
for white dwarfs is that reasonably bracket the observations. A model for a
much younger population would produce too many
1=4
T  ð2 keV=k B Þ  ð0:6 M =1:4 M Þ bright white dwarfs and not enough faint ones. A model
 ð5500 km=10 kmÞ
1=2
ð1Þ for a much older population would also fit badly, but
with the opposite signature.
or T  (70 eV/kB). Levitation is possible as the white dwarf The analogous neutron star cooling curve problem is
approaches this temperature, which is essentially the max- also now being pursued using Chandra observations of
imum possible, 800,000 K. The spectra of younger white neutron stars in SNR, a rather small sample. The two
dwarfs and those with photospheres heated by accretion problems worked together should eventually confirm
peak in the EUV. understanding of stellar evolution endpoints and evolu-
Cooling depends on the radius and hence mass and on tion of Galactic populations or else expose basic defects
composition, but the basic issue is how much heat can be in that picture. Both because there are fewer neutron
transported through the interior and radiated by the sur- stars and because their cooling models are complicated
face. Fig. 1 (Wood, 1995) shows one way of comparing by the possibility of significant non-photon contributions
cooling models with observational data. Both the data to cooling, neutron stars are probably inferior to the
(points with error bars) and the theoretical curves repre- white dwarfs for this purpose, but they do complement
sent white dwarf luminosity functions. The peak of the the white dwarf information by providing the picture
spectral energy distribution shifts from the EUV to the for a different sample with higher initial mass, where
FUV as the white dwarf cools, but the process takes a evolution goes more rapidly. Eventually, this research
significant fraction of the Hubble time. Comparison with will clarify production rates for white dwarfs, neutron
observation is made by calculating the observed distribu- stars and black holes over the life of the Galaxy. These
tion of stellar characteristics as a function of the popula- rates are important to many issues in Galactic astrophys-
tion age. The white dwarf luminosity function for the ics. Finally, the crucial data points that provide absolute
Galaxy folds together the history with which normal calibration for the whole curve by linkage to contempo-
stars have been deposited in the white dwarf stellar rary rates of white dwarf formation are the ones that
graveyard. There, they constitute an ensemble of thermal come from young, hot white dwarfs, the ones whose spec-
clocks, recording the times since endpoints of nuclear tral energy distributions peak in the EUV.
1504 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508

4. White dwarf atmospheres are needed to reconcile models and data at all wavelengths
(Barstow et al., 1999). Important progress has also been
Our understanding of physical mechanisms that deter- made in incorporating radiative levitation and diffusion
mine white dwarf evolution leaves major questions unan- self-consistently into the models (Schuh et al., 2001), where
swered. The emergence from the asymptotic giant branch the need for a He contribution is reduced but not
of two groups of white dwarfs whose compositions are eliminated.
dominated by H or He is beginning to be understood, We show in Fig. 2 the EUV spectrum of G191-B2B,
but the complex relationship between these branches and made with the J-PEX high-resolution spectrometer, flown
a demonstrable temperature gap in the cooling sequence on a sounding rocket (Cruddace et al., 2002). The model
of the He-rich branch cannot be explained. Determination had a homogeneous distribution of elements, and agree-
of the photospheric He and heavy-element content pro- ment between the model and the data is strikingly good.
vides crucial information on the evolutionary history of The broad features between 227 and 232 Å are a charac-
these stars. teristic of the overlapping series of interstellar He II
G191-B2B is one of the brightest and best studied of the absorption lines superposed on a continuum. Taken with
hot, H-rich white dwarfs, lying near the top of their cooling the strong depression in flux below 227 Å, this is strong
sequence. It is one of a group of white dwarfs with temper- evidence for the presence of interstellar or circumstellar
atures in excess of 50,000 K that contain significant quan- He II along the line of sight. Conclusive proof is
tities of heavy elements in their atmospheres, in obtained when NHeII is set to zero, and the fit degrades
particular C, N, O, Si, S, P, Fe, and Ni (Bruhweiler and markedly. The exposure was insufficient to detect any
Kondo, 1983; Sion et al., 1992; Vennes et al., 1992, 1996; photospheric He line (e.g., 243 Å) with high significance;
Holberg et al., 1994). Such material strongly depresses however, when NHeII and nHe are fitted jointly the confi-
the EUV continuum, when compared to that of stars with dence contours support a positive detection of photo-
pure H atmospheres. G191-B2B remains an important tar- spheric He. Surprisingly, models with stratified
get for EUV spectroscopy to determine the primary sources distributions have not produced better fits (Barstow
of opacity and to obtain a self-consistent model, with an et al., 2005). This example, based on only a few hundred
effective temperature, surface gravity and composition that seconds of data, marks in principle the beginning of a
can match the optical, FUV, and EUV observations new epoch of EUV astronomy, one where high spectral
simultaneously. resolution is used to disentangle source and ISM effects
Initial attempts to model its EUVE spectrum failed to and arrive at rigorous astrophysical results that are
reproduce either the observed flux level or the continuum important for both those sources and the ISM. It illus-
shape (Barstow et al., 1996), which was believed to result trates how even a small bandwidth (220–245 Å) that hap-
from inclusion of an insufficient number of Fe and Ni lines. pens to be rich in features can, when expanded with high
Adding some 9 million predicted lines to the few thousand resolution, become a field where elusive prizes concerning
with measured wavelengths did provide a self-consistent composition details may be won.
model able to reproduce the EUV, UV and optical spectra
(Lanz et al., 1996). However, good agreement required a 5. Unification of spectral and temporal regimes: CVs
significant quantity of He, either in the stellar photosphere
or in an ionized interstellar–circumstellar component. One may view white dwarfs as populating the intersec-
Unfortunately, this inferred He contribution could not be tion of two astrophysics paradigms as represented in
resolved using EUVE from the many Fe and Ni lines. Fig. 3. The additional EUV channel and the relatively long-
More recently, it has been shown that photospheric er dynamical timescales combine so that white dwarfs can
heavy elements may not be distributed homogeneously be studied in ways unavailable for neutron stars. These
(by depth) and that more complex stratified structures methodologies then overlap what can be done in laborato-

Fig. 2. EUV spectrum of the white dwarf G191-B2B made with the J-PEX instrument flown on a sounding rocket. The data are the error bars and the
best-fit model is the solid line.
 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508 1505

Fig. 3. White dwarfs are at the intersection of two paradigms in astrophysics.

ry and coronal plasma research. There is great potential non-magnetic cases separately, but making somewhat par-
here, as yet somewhat unrealized because of the demands allel points about how the EUV can contribute.
of EUV work.
CVs, which are binaries consisting of a white dwarf and 5.1. Magnetic CVs
a companion in sufficient proximity to produce a high rate
of mass transfer, are illustrative of this intersection. Exam- Consider first AM Her stars (polar), which have the
ining cases will show how timing and spectroscopy cover largest magnetic moments among compact stars. They lack
complementary aspects of these variable objects. It will also accretion disks because the magnetic field constrains flow
uncover a more negative point regarding unrealized com- all the way from the companion star to the compact object.
pleteness of understanding: there has not emerged a suc- The white dwarf and companion are locked in synchronism
cessful theoretical unification of the information derived so that the spin and orbit periods are the same. QPO phe-
from these two methodologies. For example, quasiperiodic nomena have long been known in the optical, predating
oscillations (QPOs) are a well-studied phenomenon for discovery of horizontal branch oscillations in neutron
which there is a body of theory developed by many groups, stars, the latter being the earliest NS QPOs discovered.
treating them as hydrodynamic modes produced in accre- AM Her stars have been modeled (in a succession of treat-
tion. Spectroscopy measurements bear on the state of the ments done by several groups, and gradually improving
very same accreting plasma, its density, temperature, ioni- over time in fidelity of treatment) as a magnetically fun-
zation states, and ultimately geometries. The need for uni- neled supersonic flow impacting on the white dwarf sur-
fied treatment is evident but full challenge has not been face, flowing through a standoff shock (Fig. 4) near the
taken up. Communities of theorists have specialized in surface. The shock may be unstable to an oscillatory
one or the other aspect. One can try to excuse this on the behavior, useful for diagnostic purposes. Vertical oscillato-
ground that instruments delivering spectra have not usually ry movement of the shock front is thought to manifest itself
been able to detect the temporal effects and vice versa, but as a QPO in at least two components of the three-compo-
it is also partly because of substantial difficulties. Each nent spectrum, the cyclotron (optical) emission and the (X-
body of information places strong demands on theory. ray) Bremsstrahlung. The optical oscillations are well
Those generalities apply to two distinct categories of established but the predicted X-ray QPOs are faint and
CVs. In the magnetic CVs (polars), the QPOs and associat- have not yet been confirmed. Moreover, since they derive
ed emission spectroscopy both pertain to regions near from the same accretion column the two kinds of oscilla-
polar caps. In these binary systems there is no accretion tions are predicted to be synchronized cycle by cycle, a fact
disk. In contrast, there are non-magnetic CVs where QPOs that can in principle be brought out by cross-correlating
and emission phenomena are both associated with disks. In simultaneous optical and X-ray data streams. Whether
both cases, theory has treated timing and spectra separate- the third spectral component, a 40-eV black body from
ly, despite their pertaining to the same flows. It may even- the polar cap, is also synchronized is at this time unclear
tually emerge that treatments that serve to unify one case from either the observational or the theoretical standpoint.
also help the other. Also, there is a cross-cutting theoretical Modeling of the shock is carried out primarily by means of
challenge to be sought, this being links across the popula- two-temperature numerical codes with appropriate physics
tions – either linking accreting WD to NS and BH as in for radiation and electron–ion interactions (Wolff et al.,
the schema of Table 1 or more ambitiously and elusively, 1989; Wood et al., 1992). Linear (analytical) analysis shows
linking different CV classes, with and without disks. This reduced ion–electron exchange induces the oscillation
development is also in early stages. Hence, to deal with instability, while non-linear (numerical) analysis shows that
the state of the art in CVs we must treat magnetic and thermal conduction cannot dampen the oscillations. These
1506 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508

spectral measurements (Kowalski et al., 2005) will require

high resolution (R  10,000) and sufficient sensitivity to
provide adequate statistics on dynamical timescales,
implying areas >20 cm2.

5.2. Non-magnetic CVs

Many QPO modes are known in accreting neutron

stars, analogous in the sense of Table 1 to cataclysmic
variables. Rich phenomenology can occur in a single
source, for example, Sco X-1. QPOs have provided the
best diagnostics for the low-mass X-ray binaries, where
coherent periods are both rare and transient. White
dwarf QPOs connect phenomenologically to NS QPOs,
but EUV spectroscopy is the supplementary, powerful
tool effectively available for the WD alone in determining
plasma conditions. The hope is to understand QPO phe-
nomena thoroughly in white dwarfs and then export that
understanding to neutron stars. The first goal is to
understand QPO frequencies and they vary with bright-
ness, which correlates with accretion rate. A correlation
between the low- and high-frequency QPO frequencies
exists from white dwarfs through neutron stars to black
holes. It was developed first observationally and then
Fig. 4. Schematic representation of a polar white dwarf. The accretion is explored theoretically (Belloni et al., 2002; Mauche,
modeled as a flow strictly following rigid field lines, becoming supersonic 2002a; Titarchuk and Wood, 2002; Woudt and Warner,
and then striking the star to produce a standoff shock. R* is the radius of 2002). The empirical relation, fitted in a manner guided
the white dwarf and h is a polar angle measured with respect to the by the theory, is mlow = 0.081 mhigh. In terms of the theo-
magnetic axis.
ry, both QPOs are again hydrodynamic modes, the latter
associated somewhat straightforwardly with the Kepler
frequency and the former involving magnetoacoustic
models represent the mature development of the subject effects between the field and disk. Under this picture,
and the papers cited may be used to trace back through Titarchuk and Wood (2002) derive the proportionality
the full history of this line of theory, which is not our pur- constant linking the two frequencies mMA = CMAmK,
pose here. Rather, we now turn to the disconnect between where the constant relates to geometry and the usual
this timing work and the high-resolution EUV spectrosco- plasma beta factor as:
py on the same plasmas.
p 1=2
The same accretion geometry has been explored with C MA ¼ 2 4p½ðf þ bÞ=ð1 þ bÞ ðH =Rout Þ. ð2Þ
complementary spectroscopic techniques, using EUV spec-
tra obtained at various phases of the period. An example They find that H/Rout = 0.015 and b = 0.1. These are re-
using EUVE is the phase-resolved spectrum of AR UMa cent developments, but they show the enticing possibility
(an AM Her-type or polar system) obtained by Szkody of a QPO phenomenon that actually bridges the gap be-
et al. (1999). These systems, then, are simultaneously rich tween white dwarfs and neutron stars.
in both timing and spectroscopic information and each Once again, we turn from the timing theory at a rep-
body of observational knowledge has spawned correspond- resentative state of development and turn to the comple-
ing – yet independent – theory. Not only is the theoretical mentary EUV spectroscopy. This field is in too great a
synthesis awaited, the observational picture also falls far state of flux for it to be the case that the timing and
short of what one would like, namely simultaneity of cov- spectroscopy are invariably available on the same
erage in optical, EUV, and X-rays, in which the temporal objects, but nonetheless a major mystery is emerging
information reaches the known dynamical timescales (sec- on the spectroscopic side. EUVE spectra of some dwarf
onds) and samples the phases of the period, while the spec- novae are rich in features, which are persistent but not
troscopy harvests the plasma diagnostics from the lines, all well understood (Mauche, 2002b). The intermediate polar
simultaneously. However, the EUVE mission has ended, EX Hya is a non-magnetic CV and also a strong EUV
CHIPS (Cosmic Hot Interstellar Plasma Spectrometer) is source for which the spectral information is unusually
optimized for diffuse emission, and while the Chandra res- good. Belle et al. (2002) shows the time-resolved spectra
olution is good at EUV wavelengths (R  2000) its sensitiv- from EUVE. The number of significant line detections is
ity is low. No new missions are planned, but future EUV small and represent blends of Fe and other elements. The
 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508 1507

first challenge here is to unravel these spectra, identifying the case as high resolution EUV spectroscopy begins to
the features with confidence, so that the problem may contribute significantly.
yield to higher resolution.
7. Conclusions
6. Cosmological issues
White dwarfs are a key component of the compact pop-
The EUV viewpoint on compact objects continues
ulations representing stellar evolution endpoints. Under-
beyond the Galaxy to extragalactic astrophysics and cos-
standing their phenomenology and sub-populations in an
mology. Some of this again involves white dwarfs indirect-
evolutionary context, particularly if pursued in conjunction
ly, while other aspects involve AGN. First, type Ia
with a similar effort on neutron stars, is central to the astro-
Supernovae are now accepted as the chief standard candle
physics of compact objects. In recent decades X-ray diag-
of the cosmic distance scale, which led to the original (and
nostics have been used to advance dramatically the
still most direct) basis for inferences regarding the existence
characterization and understanding of the neutron star
of Dark Energy. Missions such as JDEM, the Joint Dark
population, but progress on white dwarfs has been slowed
Energy Mission (for which one candidate realization is
by lack of suitable photometry and spectroscopy at EUV
SNAP, the Supernova Acceleration Probe), will observe
wavelengths, despite the fact that they are comparatively
large numbers of these SNe, class by class, out to
numerous so that there are outstanding white dwarf ‘‘lab-
z = 1.7. The assumed type Ia progenitor is a carbon–ox-
oratories’’ in the Solar neighborhood. It is necessary to
ygen core white dwarf, which has exceeded its Chandrase-
achieve sufficient spectral resolution to permit study of
khar limit at 1.4 Mx, either through accretion or merger of
the white dwarfs through the absorption and emission fea-
a double-degenerate system. This implies the existence of a
tures of the ISM. This obstacle is rapidly being overcome.
population of short period binary systems that will evolve
Suitable technologies have progressed to the point where
into type Ia events within a cosmological timescale (Gyr).
practical photometers and spectrometers can be designed
Unfortunately, in spite of several comprehensive searches,
with effective areas >20 cm2 and with spectral resolution
few clear examples of such systems have emerged. Prudence
10,000 (Kowalski et al., 2003a,b).
may favor getting better observational understanding of
such precursor systems as a prerequisite to the develop-
ment of such large missions. As the progenitors are likely References
to be hot but scarce, deeper white dwarf EUV surveys with
Barstow, M.A., Hubeny, I., Lanz, T., Holberg, J.B., Sion, E.M. The
high-resolution spectroscopy followup may discover composition and structure of white dwarf atmospheres revealed by
candidates. extreme ultraviolet spectroscopy, in: Bowyer, S.C., Malina, R.F.
Second, much emphasis in NASA’s Strategic Plan and (Eds.), IAU Colloq. 152, Astrophysics in the Extreme Ultraviolet.
SEU/Origins roadmaps has been given to the investigation Kluwer, Dordrecht, pp. 203–210, 1996.
Barstow, M.A., Hubeny, I., Holberg, J.B. Evidence for the stratification of
of the early universe. Sources at cosmological distances
Fe in the photosphere of G191-B2B. Mon. Not. Roy. Astron. Soc. 307,
have spectral energy distributions redshifted, which then 884–894, 1999.
sets the bandwidth requirement for corresponding instru- Barstow, M.A., Cruddace, R.G., Kowalski, M.P., Bannister, N.P., Yentis,
mental designs. A striking illustration of this is the fact D., Lapington, J.S., Tandy, J.A., Hubeny, I., Schuh, S., Dreizler, S.,
the James Webb Space Telescope (JWST) will follow up Barbee, T.W. High resolution EUV spectroscopy of G191-B2B II:
cosmological work initiated in the visible and near-UV structure of the stellar photosphere and the surrounding interstellar
medium. Mon. Not. Roy. Astron. Soc. 362, 1273–1278, 2005.
wavelengths using HST, but does so with an instrument Belle, K.E., Howell, S.B., Sirk, M.M., Huber, M.E. An extreme-
optimized for infrared wavelengths – the UV science pur- ultraviolet study of the intermediate polar EX hydrae. Ap. J. 577,
sued to higher z moves to the IR. Similarly, X-ray sources 359–370, 2002.
at z > 5 (i.e., AGN) will find shifted into the ultrasoft X- Belloni, T., Psaltis, D., van der Klis, M. A unified description of the timing
features of accreting X-ray binaries. Ap. J. 572, 392–406, 2002.
ray/EUV a portion of their output that increases as z
Bruhweiler, F.C., Kondo, Y. Mass loss, levitation, accretion, and the
increases, which is to say as they become more interesting sharp-lined features in white dwarfs. Ap. J. 269, 657–667, 1983.
cosmologically. This includes, for example, the spectral Cruddace, R.G., Kowalski, M.P., Yentis, D., Brown, C.M., Gursky,
regions near K transitions of C, N, and O, and L-transi- H., Barstow, M.A., Bannister, N.P., Fraser, G.W., Spragg, J.E.,
tions of Fe at rest energies 1 keV. Although extragalactic Lapington, J.S., Tandy, J.A., Sanderson, B., Culhane, J.L., Barbee,
T.W., Kordas, J.F., Goldstein, W., Fritz, G.G. High-resolution
measurements are complicated by interstellar absorption at
spectroscopy of G191-B2B in the extreme ultraviolet. Ap. J. 565,
wavelengths above 100 Å, regions (e.g., the Lockman hole) L47–L50, 2002.
of low column encompass sufficient solid angle to provide Eddington, A.S. Internal Constitution of the Stars. Cambridge University
reasonable numbers of extragalactic targets for instruments Press, Cambridge, 1926.
with sufficient sensitivity. Moreover, accretion processes Holberg, J.B., Hubeny, I., Barstow, M.A., Lanz, T., Sion, E.M., Tweedy,
R.W. The discovery of Ni V in the photospheres of the hot DA white
ongoing in distant X-ray AGNs are self-similar to those
dwarfs RE 2214-492 and G191-B2B. Ap. J. 425, L105–L108, 1994.
in CVs. There has been continual borrowing of ideas and Lanz, T., Barstow, M.A., Hubeny, I., Holber, J.B. A self-consistent
insights from the Galactic accreting sources to the AGN optical, ultraviolet, and extreme-ultraviolet model for the spectrum of
and vice versa and one should anticipate that this will be the hot white dwarf G191-B2B. Ap. J. 473, 1089–1093, 1996.
1508 K.S. Wood et al. / Advances in Space Research 38 (2006) 1501–1508

Kowalski, M., Barstow, M., Cassinelli, J., Cruddace, R., Doschek, G., Schuh, S., Dreizler, S., Wolff, B. Application of stratified non-LTE
Dupree, A., Guinan, E., Howell, S., Linsky, J., Livio, M., Malina, R., model atmospheres to hot DA white dwarfs, in: Provencal, H.L.,
Sarazin, C., Siegmund, O., Stern, R., Walter, F., Werner, K., Wood, Shipman, J., Macdonald, Goodchild, S. (Eds.), ASP Conf. Ser. 226,
K., Welsh, B., Wilkinson, E. The future of EUV astronomy, topical 12th European Workshop on White Dwarfs. ASP, San Francisco, pp.
session and white paper, in: L. Strickman, D. Yentis, M. Kowalski 79–85, 2001.
(Eds.), Proceedings, Naval Research Laboratory: Washington, DC. Sion, E.M., Bohlin, R.C., Tweedy, R.W., Vauclair, G.P. The identification
Publication NRL/MR/7650-04-905, 2005. of highly ionized metal species in the hot DA1 white dwarf G191-B2B
Kowalski, M.P., Cruddace, R.G., Wood, K.S., Yentis, D.J., Gursky, H., with the Hubble Space Telescope Faint Object Spectrograph. Ap. J.
Barbee Jr., T.W., Fritz, G.G., Hunter, W.R., Heidemann, K.F., 391, L29–L32, 1992.
Barstow, M.A. Proposed multilayer-grating designs for the Astro- Szkody, P., Vennes, S., Schmidt, G.D., Wagner, R.M., Fried, R., Shafter,
physical Plasmadynamic Explorer (APEX): an EUV high resolution A.W., Fierce, E. ASCA, RXTE, EUVE, and Optical Observations of
spectroscopic SMEX, in: Citterio, O., O’Dell, S.L. (Eds.), Optics for the High Magnetic Field Cataclysmic Variable AR Ursae Majoris. Ap.
EUV, X-ray, and Gamma-ray astronomy, SPIE Proc. 5168, pp. 21–30, J. 520, 841–848, 1999.
2003a. Titarchuk, L., Wood, K. On the Low and High Frequency Correlation in
Kowalski, M.P., Cruddace, R.G., Wood, K.S., Yentis, D.J., Wolff, M.T., Quasi-Periodic Oscillations among White Dwarf, Neutron Star, and
Laming, J.M., Gursky, H., Carruthers, G.R., Barbee Jr., T.W., Kordas, Black Hole Binaries. Ap. J. 577, L23–L26, 2002.
J.F., Mauche, C.W., Fritz, G.G., Varlese, S.J., Barstow, M.A., Fraser, Vennes, S., Chayer, P., Hurwitz, M., Bowyer, S. Discovery of photo-
G.W., Siegmund, O.H.W., Welsh, B.Y., Brickhouse, N.S., Dupree, spheric sulfur and phosphorus in ORFEUS spectra of the hot white
A.K., Brown, A., Bruhweiler, F.C., Cameron, A.C., Holberg, J.B., dwarfs G191-B2B and MCT 0455-2812. Ap. J. 468, 898–906, 1996.
Howell, S.B., Jordan, C., Linksy, J.L., Matthews, S.A., Sion, E.M., Vennes, S., Chayer, P., Thorstensen, J.R., Bowyer, S., Shipman, H.L. Iron
Werner, K. Proposed mission concept for the Astrophysical Plasma- abundance in the hot DA white dwarfs Feige 24 and G191 B2B. Ap. J.
dynamic Explorer (APEX): An EUV high resolution spectroscopic 392, L27–L30, 1992.
SMEX, in: Siegmund, O.H.W. (Ed.), UV/EUV and Visible Space Wolff, M.T., Gardner, J.H., Wood, K.S. On the stability of degenerate
Instrumentation for Astronomy II, SPIE Proc. 5164, pp. 1–16, 2003b. dwarf radiative shocks: new one- and two-temperature calculations.
Mauche, C.W. Correlation of the quasi-periodic oscillation frequencies of Ap. J. 346, 833–843, 1989.
white dwarf, neutron star, and black hole binaries. Ap. J. 580, 423–428, Wood, K.S., Imamura, J.N., Wolff, M.T. Noise-driven tadiative shocks.
2002a. II. Further implications for the quasi-periodic oscillations in AM
Mauche, C.W. EUVE observations of nonmagnetic cataclysmic variables, herculis objects. Ap. J. 398, 593–601, 1992.
in: Howell, S.B., Dupuis, J., Golombek, D., Walter, F.M., Cullison, J. Wood, M.A., 1995. Private communication after Wood, M.A., in:
(Eds.), Continuing the Challenge of EUV Astronomy: Current Koester, D., Werner, K. (Eds.), White Dwarfs. Lecture notes in
Analysis and Prospects for the Future, ASP. Conf. Ser. 264. ASP, Physics 443, pp. 41, 1995.
San Francisco, pp. 75–86, 2002b. Woudt, P.A., Warner, B. Dwarf nova oscillations and quasi-Periodic
Oppenheimer, J.R., Volkoff, G.M. On Massive Neutron Cores. Phys. Rev. oscillations in cataclysmic variables: I. Observations of VW Hyi. Mon.
55, 374–381, 1939. Not. Roy. Astron. Soc. 222, 411–422, 2002.