I provide examples of how high-resolution x-ray spectra may be used to determine the temperature and emission measure distributions, electron densities, steady and transient flow velocities, and location of active regions in stellar coronae. For each type of measurement I estimate the minimum spectral resolution required to resolve the most useful spectral features. In general, high sensitivity is required to obtain sufficient signal-to-noise to exploit the high spectral resolution. Although difficult, each measurement should be achievable with the instrumentation proposed for AXAF.

X-ray spectra of stellar X-ray sources taken with high, moderate and low spectral resolution are discussed. Only low resolution spectra are available for a sufficiently large number of late-type stars. It is shown that high temperature plasmas (log T > 7) produce the dominant emission component in the coronae of red dwarfs as well as yellow giants, while the coronae of F stars – usually – have X-ray temperatures similar to those found in the quiet Sun. The need for density diagnostics of stellar coronae is stressed and it is argued that the waveband region between 90 – 140 Å is particularly well suited for this purpose.

We summarize some of the results obtained by spectral analysis of EXOSAT ME observations of stellar coronal sources. We focus on time-resolved spectroscopy of stellar flares and on the determination of the temperature structure of quiescent RS CVn binaries

Compact solar flares are triggered by sudden energy release in magnetically confined plasma. This class of flares is well suited to be studied with numerical hydrodynamic models. In particular, one can compare the evolution of observed and synthetic X-ray spectra, computed under various assumptions for the mechanism of impulsive energy deposition, to constrain theoretical models and their parameter space. We discuss recent results on solar flares along this line, non thermal to models of energy depositions by relativistic electron beams. We shall also discuss possible applications of X-ray spectral synthesis to stellar flares.

We discuss theoretical X-ray spectra for coronal loop models in the rapidly rotating young K dwarf AB Doradus (HD 36705), as a typical representative of active X-ray bright stars. The loop models are based on EXOSAT and IUE observations, and further motivated by a possible connection between the observed X-ray flares and co-rotating clouds of neutral hydrogen (a few/day). The resulting synthetic spectra between 0.5 – 7 keV can be approximated by a linear combination of three distinct temperature components. Two components are sufficient between 0.5 – 2.5 keV. Below 0.1 keV the loop spectra deviate significantly from the few component fits. To test some basic assumptions (dynamic vs. static, constant vs. variable cross-sectional area), useful constraints on the DEM(T)-distribution could be obtained with the grating-spectrometer of XMM with 103 – 104sec exposure times. The ratio of the He-type (O VII) forbidden and intercombination lines at 0.56 keV will provide sufficient density diagnostics, to distinguish e.g.between compact and large loops. The crystal-spectrometers of XMM and XSPECT could achieve the same but with longer (105 sec) observing time. The strong Ly α line of O VIII at 0.65 keV can be observed with the crystals in 104 sec, and used even for rotational modulation and flare studies, and giving additional information about flows in flaring loops. At the Iron 6.7 keV lines, where the gratings do not work, the crystals should be used together with low resolution devices, to set constraints on the hottest gas at loop summits. Our discussion applies also to several brighter cool stars but with shorter observing time (like Capella and HR 1099, which are over 3 times X-ray brighter than AB Dor).

The MIT group has used data from the Focal Plane Crystal Spectrometer on the Einstein Observatory to perform plasma diagnostics of four supernova remnants (SNRs), the Cygnus Loop, Puppis A, N132D, and Cas A. The ratio of luminosities of the forbidden line to resonance line of He-like ions of oxygen and neon allow us to show that all four SNRs depart from ionization equilibrium in that they are under-ionized for their electron temperatures. Thus despite the fact that their ages range from 300 yr to 20,000 yr, all four SNRs are still ionizing and, in that sense, are still young. We derive values of ionization time and electron temperature for one or more components in each remnant. The agreement between these values and those deduced by others using entirely different means (e.g. broad-band spectroscopy or imaging) gives us confidence in the reliability of the diagnostics. Two of the SNRs, Puppis A and N132D, show evidence for an overabundance of oxygen by factors of three or more. These results, based on a handful of weak lines, show the great promise of the much more powerful future spectroscopy missions for revealing detailed information about astrophysical plasmas.

The soft X-ray background is explained in terms of emission coming from hot gas. Most of these soft X-ray data were obtained by proportional counters with a poor energy resolution. Instruments having the capability to resolve lines were only flown by two groups: a GSPC by a Japanese group and a SSD by a french-american collaboration. They both detected the 0 VII line emission coming from the soft X-ray background and so proved the thermal nature of the emission. The implications of these results on possible models for the local hot medium will be discussed. The same detectors observed part of the North Polar Spur. They detected emission lines coming from different species (0 VII,Fe XVII,Ne IX). Spatial variations of line ratios for this object could be due to non-equilibrium ionization effects.

We present a detailed numerical study of the X-ray emission of Tycho’s supernova remnant. Using Nomoto’s W7 model of an exploding white dwarf as intial condition we follow the hydrodynamical and ionisation evolution for different densities of the interstellar medium and compare the emitted spectra with EXOSAT GSPC observations. The results indicate that the density of the ambient medium is 0.5 ≤ no ≤ 1.0 cm-3, the distance to Tycho is about 3 kpc.

We discuss pulse height analysis of Be band data in relation to the important 72 eV Fe line cluster emission from the soft X-ray background (SXRB). Pulse height fits to the Be band data suggest that the Fe lines must be suppressed by a factor of ~10 with respect to the rest of the X-ray spectrum. The broad band rates and the mean energy of the pulse height data for the Be band can be brought into agreement by using depleted elemental abundance emission models. A planned measurement of the SXRB Fe lines using the Array of Low Energy X-ray Imaging Sensors (ALEXIS) experiment could resolve this issue.

A simple physical model of the dynamics of a young supernova remnant is used to derive a straightforward kinematical description of the reverse shock. With suitable approximations, formulae can then be developed to give the X-ray emission of the reverse-shocked ejecta. The results are found to agree favorably with observations of SN1006.

A weak linear correlation is found between the x-ray and radio surface brightnesses of galactic supernova remnants. A flatter slope is found for the infrared-radio correlation.

High resolution X-ray spectral observations of Puppis A were performed with the FPCS on the Einstein Observatory at three regions of the remnant: the shock front, the bright eastern knot, and the interior. Plasma diagnostics of lines from OVII and OVIII constrain the values of electron temperature, ionization timescale, and hydrogen column density. We compare results of the diagnostics for these three regions. A non-equilibrium analysis of previously published fluxes of oxygen lines shows that the interior has not yet reached ionization equilibrium.

The eclipsing RS CVn system AR Lac (G2IV+K0IV) has been observed continuously for one 2 day binary cycle with the EXOSAT observatory. Below 1 keV a factor of two intensity modulation is seen with a flat bottomed minimum around the time of primary eclipse and a shallow dip preceeding primary eclipse. Above 1 keV, where only emission with temperatures > 107 K would be detected, no orbital modulation or eclipse is seen. This suggests that the > 20 million degree emission comes from a large region, comparable in size to the binary seperation. The modulation in the < 1 keV lightcurve has been modelled by x2 fitting to X-ray bright spots and by using maximum entropy deconvolution. The lower temperature emission is found to orginate in compact regions with a pressure and temperature similar to that of the flaring sun.

Accretion disk coronae are likely to be the dominant site for X-ray absorption and reprocessed emission in low mass X-ray binaries, and may be present in other classes of compact X-ray sources such as active galactic nuclei and cataclysmic variables. In spite of this fact, and in spite of the observational evidence for their existence, there remain many uncertainties about the structure of accretion disk coronae. This paper will discuss the coronal structure and dynamics, their X-ray spectral signatures including coupling to the variability behavior of compact X-ray sources, and the major unsolved theoretical issues surrounding them.

Cygnus X-2 was observed with EXOSAT at five phases of a single orbital cycle in September of 1983. We will summarize the results of spectral fits of the LE + ME (Argon) data in terms of a superposition of thermal bremstrahlung and blackbody components. During the first observation a grating spectrum was obtained and this is described in some detail. The GSPC data are used to investigate the presence of iron features, and their behavior during dips.

We observed the 3.41 day eclipsing, massive binary system 4U1700-37/HD153919 with EXOSAT for more than one complete binary period to investigate the spectral variations during the orbital cycle of the neutron star. The spectra show a low energy excess below ~ 3 keV when modelled by a powerlaw spectrum attenuated by photoelectric absorption by neutral gas, suggesting partial ionization of the absorbing gas. The column density derived from spectra above 3 keV shows an asymmetric distribution around orbital phase 0.5 with higher absorption before eclipse ingress. We approximated the with distance to the X-ray source gradually decreasing ionization of the wind by two zones. One of higher ionized wind around the X-ray source for which X-ray opacities of a gas in photoionization equilibrium were used and a zone of neutral gas further away from the X-ray source. We find that our spectra below 3 keV can be well fitted by a powerlaw which is attenuated first by photoelectric absorption of ionized gas and then by neutral gas. Since around phase 0.5 the major contribution of the wind column density along the line of sight arises from the ionized part we found that the total column density can be higher up to a factor of about 4 taking ionization into account.

The X-ray transient 4U1543-47 was observed in 1983 by the EXOSAT observatory near the maximum of an outburst. The X-ray spectrum was measured using a gas scintillation proportional counter (GSPC) and a transmission grating spectrometer (TGS). Two emission line features are resolved. A broad (FWHM ~2.7 keV) line at 5.9 keV is detected in the GSPC, which we interprete as a redshifted and broadened iron Kα line. The Une broadening and redshift may arise from either Compton scattering in a cool plasma with small optical depth (τ ≈ 5), or from Doppler and relativistic effects in the vicinity of a compact object. The spectrum below 2 keV, obtained with the TGS, shows evidence for a broad emission line feature at 0.74 keV, which may be an iron L-transition complex. However, we find that such an emission feature could be an artifact caused by an anomalously low interstellar absorption by neutral Oxygen. The continuum emission is extremely soft and is well described by an unsaturated Comptonized spectrum from a very cool plasma (kT = 0.84 keV) with large scattering depth (τ ≈ 27). The continuum spectrum is strikingly similar to that of black hole candidate LMC X-3.

X-ray line observations of clusters of galaxies have shown that the X-ray emission in clusters is mainly thermal emission from hot diffuse gas, and that much of this gas has come out of stars, probably having been ejected from galaxies in the cluster. Future high resolution observations should allow us to determine the physical state of the gas. X-ray line measurements and abundance determinations can lead to strong constraints on the origin of the intracluster gas, and on the chemical evolution and history of galaxies. Some of the stronger resonant X-ray lines may be observable as absorption lines against a background quasar. Such X-ray absorption line measurement can be used to directly derive distances to clusters, using a technique similar to (and possibly complementary to) that the well-known method using the Zel’dovich-Syunyaev effect.

The evidence for cooling flows in clusters of galaxies is discussed. Peaked X-ray surface brightness profiles and the presence of soft X-ray spectral components are characteristic signatures of cooling flows. The best available data are consistent with gas cooling at a rate of ~ 100M⊙yr-1 and depositing cooled matter over a radius of 100 – 200 kpc. The same rate is obtained from observations of gas cooling over a range of temperatures and thus a range of cooling times covering less than 3 × 107 yr to more than 3 × 109 yr. It appears that cooling flows are both steady and long-lived. Some indirect optical evidence for distant cooling flows (0.5 < z < 1) is presented.

Arguments against the existence of large scale cooling flows in clusters of galaxies are presented. The evidence for cooling flows is all circumstantial, consisting of observations of cool gas or hot gas with a radiative cooling time less than the Hubble time, or a central peak in the x-ray surface brightness profile. There is no evidence for large quantities (several tens to several hundreds of solar masses per year) of matter actually flowing anywhere. On the contrary, several lines of evidence — stellar dynamics, observations of the amount of star formation, x-ray surface brightness observations, theoretical calculations of the growth of thermal instabilities, the amount of cold gas — suggest that cooling flows, if they exist, must be suppressed by one to two orders of magnitude from the values implied by simple estimates based on the radiative cooling time of the x-ray emitting gas. Two heat sources which might accomplish this — thermal conduction and relativistic particles, are considered and an alternative to the standard model for cooling flows is presented: an accretion flow with feedback wherein the accretion of gas into a massive black hole in the central galaxy generates high energy particles that heat the gas and act to limit the accretion.