Evidence has been accumulating that the gas which produces the broad line emission (BLR gas) in active galaxies has a range of physical conditions, and may indeed come from a distribution of spatially distinct regions (Gaskell 1982, Ulrich et al. 1984, Netzer 1987, Collin-Soufrin et al. 1988). In this paper, we try to separate out any contribution to the BLR emission line profiles from optically thin material. We do this by comparing the profiles of OI λ8446 (a flourescence line only produced by clouds very optically thick to Lyman α, Elitzur and Netzer 1985) and Hα (which can be produced by optically thick and optically thin clouds).

Many scenarios of the evolution of star clusters in the centers of galaxies involve the formation of a central supermassive object. Since black hole formation is not 100% efficient in mass usage, stars are left over. This paper investigates the properties of such stars and proposes that their externally heated atmospheres become “bloated” due to radiative forces from trapped line radiation. Such stars would swell to many times their normal diameters and acquire densities, sizes, and mean column masses typical of QSO/AGN emission line clouds (ELCs).

The spectra of active galactic nuclei (AGNs) are characterised by broad permitted emission lines. The relative intensities of these lines are roughly similar in spectra of galaxies of very different continuum luminosities. The most likely source of energy for the line emitting gas is photoionization by an energetic (non-stellar) radiation field.

I discuss physical processes which may be important in the central engines of AGNs. Black hole accretion models have been elaborated during the past few years, and provide a plausible framework for jet and spectrum formation. However, the specific mechanisms responsible for producing the observed continuum radiation remain uncertain. I consider nonthermal models for the IR continuum, constraints on thermal plasma, and the production of relativistic electrons.

Results are presented from EXOSAT observations of Seyfert type active galactic nuclei (AGN). The sample chosen for study are 48 hard X-ray selected Seyfert type AGN, including all 30 of the emission line AGN in the Piccinotti sample. Combining EXOSAT LE and ME data has allowed us to obtain broad band X-ray spectra over the range 0.1–10 keV. Spectra in the ∼ 2–10 keV range are found to be well described by a simple power-law. The distribution of spectral indices across the sample can be approximated by a Gaussian distribution of mean energy index ∝ = 0.70 with σ∝ = 0.17 although not all individual spectra are consistent with this mean at the 90% confidence level. EXOSAT has also revealed a substantial number of sources with complex soft X-ray spectra. These include spectra with a second spectral component at soft X-ray energies and sources with “leaky” absorbing columns. Evidence that soft excess components occur in at least 50% of Seyfert type AGN together with detection of rapid variability in the soft component provides quantitative support for an accretion disc model for AGN.

We present mean spectral parameters for various ensembles of quasars observed with the Einstein Observatory Imaging Proportional Counter (IPC). Our sample contains 71 optically or radio selected quasars with 0.1 < z < 3.5, Galactic NH < 1021 cm−2, total counts of 30 −500, and IPC gain < 19. Quasars are grouped into ensembles according to radio properties (Flat Radio Spectrum [FRS], Steep Radio Spectrum [SRS] or Radio Quiet [RQ]), and either redshift or X-ray luminosity, lx. We find a clear correlation between radio properties and α. FRS quasars have α∼0.4, SRS quasars have α∼0.7 and RQ quasars have α ∼1–1.4. There is no evidence for a dependence of α on z nor, for the FRS and SRS ensembles, on lx over nearly three decades. FRS quasars with 2.0 < z < 3.5 have just as flat mean spectra as those with low z, implying that a single power law, which is flatter than the canonical one with α ∼ 0.65, continues into the 1–10 keV band (in which the observed softer X-rays were emitted). Unfortunately, the results for high redshift and high lx RQ quasars are ambiguous because of systematic uncertainties in the ensemble means. Thus we cannot test the two-component spectral hypothesis of Wilkes and Elvis for these objects. SRS X-ray spectra could be steeper than FRS spectra because of the mixing of two components, although a single intrinsically steeper spectrum is easier to reconcile with the absence of z dependence. The uncertainty in a for RQ quasars with high z leaves open the important question of their contribution to the cosmic X-ray background.

X-ray emission in the 2–10 keV energy range was observed with the Ginga satellite from the Seyfert 2 galaxy NGC1068. The continuum spectrum can be described by a power-law of photon index about 1.5. An intense iron line at 6.5 keV with an equivalent width of 1.3 keV was clearly noticed. The X-ray flux was about 6 × 10 −12 erg/sec/cm2 or 3 × 1041 erg/sec, assuming a distance of 22 Mpc. The observed spectrum is consistent with the scattering and reprocessing of X-rays by the gas surrounding the central engine. With this picture we estimate that the X-ray flux of the central engine is about 1043 - 1044 erg/sec, a typical value for a Seyfert 1 galaxy.

The first X-ray (2–10 keV) spectral surveys of active galactic nuclei (AGN) were remarkably uniform, with spectral indices narrowly distributed about α=0.65, independent of radio-emitting properties or X-ray luminosity (Mushotzky 1984). In the few AGN detected above 20 keV, this canonical power law extended to at least 50 keV (Rothschild et al. 1983, Pounds 1985, Worrall et al. 1980; Halpern 1982; Worrall and Marshall 1984). In practice these surveys were dominated by low luminosity Seyfert 1 galaxies.

Several lines of evidence suggest that the x-ray spectra of quasars are not simple, exact power laws: 1. when Wilkes and Elvis (1987) analyzed quasars as power laws they found an absorption less than that due to our galaxy; 2. The mean 0.3 to 3.5 keV spectral index is steeper than the mean for the 2 to 20 keV range; 3. although several lines of evidence argue that AGN provide a significant portion (perhaps all) of the x-ray background, the diffuse background spectrum does not agree with the x-ray power-law indices measured for quasars or Seyfert galaxies. Schwartz and Tucker (1988) have suggested that all the above conflicts are reconciled if the slope in the Log(flux density) vs. Log(energy) plot flattens continuously with increasing energy. In this paper we utilize one particular parameterization suggested for the flux density, which we call the “log-slope” model:

where f is the flux density, K a normalization parameter which is not of interest here, and a and b are the two parameters of our fit.

The Seyfert galaxy NGC5548 was observed with EXOSAT 12 times between February 1984 and March 1986. The bottom panel of Fig. 1 shows the lightcurves obtained with the Low Energy telescope and CMA detector and the Medium Energy proportional counters. The CMA flux doubled in a month at the beginning of the EXOSAT coverage, and was down by more than an order of magnitude 15 months later; the source was again in a high state in January and March 1986. The ME lightcurve mirrors that of the CMA, with L2–6 keV varying between 1 and 5 × 1043 erg s−1.

We observed the Seyfert galaxy NGC6814 five times with EXOSAT between September 1983 and October 1985. For a more detailed description of the data see Mittaz and Branduardi-Raymont (1988). Between observations both the LE (0.02 - 2.5 keV) and ME (2 - 6 keV) fluxes vary in the same sense, NGC6814 being brightest during observation 1 and weakest during observation 3 when it was undetected by the CMA. Both variable luminosity and variable low-energy absorption can explain this.

The medium energy X-ray spectrum of the Seyfert galaxy NGC 4151 is characterised by a hard power-law continuum, a variable low-energy cut-off and iron line and edge features. Holt et al. (1980) found the spectral form below 4 keV to be incompatible with absorption in a uniform screen of cold gas and interpreted the apparent soft X-ray excess in terms of a partial covering model. However, recent observations point to a somewhat different description of the X-ray absorbing medium present in the nucleus of this galaxy.

The first Einstein IPC X-ray spectra of quasars had slopes similar to their infrared spectra (Elvis et al. 1986) suggesting that the infrared and x-ray continua of quasars might be two ends of a single power-law extending over at least 4 decades of the electromagnetic spectrum. In retrospect this was a bold claim based as it was on only eight quasars with limited infrared coverage. If true though it would greatly simplify the structure of the quasar continuum.

Until recently there have been very few measurements of the ionizing continuum in quasars. With the combination of Einstein X-ray slopes and IUE spectra for a sample of quasars, this continuum can now be better constrained. Here we take a preliminary look for relations between the shape of this continuum and the emission lines produced over the observed range of continuum properties.

Recent years have seen some progress in the studies of the central engine of active galactic nuclei (AGN). On the observational side, past space missions such as HEAO-A, Einstein and EXOSAT have brought some considerable progress in the observations of X-ray spectra and time variabilities of AGN. More significant progress is expected in the immediate future due to the current and future space missions such as IUE, GINGA, the Hubble Space Telescope, ASTRO-D, ROSAT, and AXAF. Therefore, it would be important to continue parallel theoretical studies. In this report composite models are presented which naturally explain the observed continuum radiation from seyferts and radio-quiet quasars. Our aim is to explore the central engine of AGN. This class of AGN is selected because it is generally considered to give the most direct information on the central power-house.

A wide variety of X-ray spectral forms has been reported in BL Lac objects. Concave spectra, i.e. a steep soft X-ray spectrum with a flat high energy tail, have been reported in a few of the brightest BL Lacs (e.g Urry 1986). Conversely, convex spectra (steep hard X-rays, flat soft X-ray spectrum) have also been reported, sometimes in the same objects (Madejski 1985, Barr et al 1988, George et al 1988). The high energy tails have usually been invoked as a signature of synchrotron-self-Compton emission. Two conflicting interpretations of the convex spectra have been made. Urry et al (1986) suggest absorption by a partially ionised medium, probably intrinsic to the BL Lac object, following the identification of an Oxygen absorption trough in the Einstein OGS spectrum of PKS 2155-304 by Canizares and Kruper (1984). Conversely, Barr et al (1988) attribute the hard X-ray steepening to energy loss mechanisms operating on a synchrotron source.

The X-ray luminosity and variability of many AGN are sufficiently extreme that any hard γ-rays produced in the source will collide with the X-rays and create electron-positron pairs, rather than escape. A small region where vast amounts of energy are produced, such as an AGN, is an ideal place to accelerate particles to relativistic energies and so produce γ-rays by Compton scattering. The observed X-ray spectra of AGN are hard and indicate that most of the luminosity is at the highest energies so that absorption of the γ-rays represents a large fraction of the energy flux, which can then be re-radiated at lower energies. Pairs can thus effectively reprocess much of the radiant power in an AGN.

Quasars with similar core-compact radio properties can be classified by their differences at optical and infrared frequencies. Their X-ray properties might be expected to be similar if the synchrotron self-Compton mechanism relates their radio and X-ray emission. We have compared the 0.2–3.5 keV mean power-law energy spectral indices, , for 4 quasar classes: 12 Highly Polarized QSOs (HPQs), 19 Flat Radio Spectrum, core-compact, low-polarization, QSOs (FRS QSOs), 24 radio-selected BL Lac objects, and 7 X-ray-selected BL Lac objects.

A sample of Active Galactic Nuclei (AGN) have been discovered during a program to identify the optical counterparts of X-ray sources detected by the Modulation Collimator experiment of the High Energy Astronomy Observatory-1 (HEAO-1). UV-excess techniques were used to identify the X-ray sources (Remillard et al. 1986) and the details of the identifications are given elsewhere (Remillard et al. 1988, Brissenden et al. 1988). We report here the preliminary results of a multi-wavelength study of these new AGN.

The observed spectral index as a function of frequency of QSO continua must be explained in models. It is generally increasing (F(ν) ∝ ν−α, α increasing) with higher frequency in the infrared (downward curvature). The visual to ultraviolet continuum has been shown to be a broken power law with F(ν) ∝ ν−0.5 at low frequency and a break to larger α at νo ∼ 3×1015 Hz. X-ray observations frequently exhibit a flat continuum with α < 1. One prominent example is 3C273 for which α1–3μm → 2, αvis ∼ 0.5 and αx ∼ 0.5. These spectral indices arise naturally in Secondary Electron Synchrotron Self-Compton (SESSC) models. Some accretion disk models approach these spectral indices for the visual-ultraviolet portion of the spectral distribution.