2.1. General Characteristics
Figure 1 shows examples of fairly well-observed SEDs for both a radio-quiet (RQQ) and a radio-loud (RLQ) quasar. Both classes show peaks in their energy output in the infra-red (IR bump) and optical ("Big Blue Bump") wavebands. The IR bump is generally attributed to thermal emission from dust at a wide range of temperatures, ~ 50 -1000 K and the Big Blue Bump in thermal emission from the gas in an accretion disk. The relative strengths of the IR and Big Blue bumps varies but they are generally comparable. The inflection between the two peaks, at ~ 1.5 µm, is due to the maximum dust temperature of ~ 2000K caused by sublimation (Sanders et al. 1989). In the X-ray region, ~ 50% of both RLQs and RQQs have a soft X-ray excess component thought to be the high energy tail of the Big Blue Bump.
Figure 1. Radio - X-ray spectral energy distribution of radio-loud (upper) and radio-quiet (lower) quasar, Elvis et al. (1994).
At harder X-ray energies, power law emission has differing slopes and relative strengths. RQQs typically have slope, E ~ 1.0 ± 0.5 (where F -E, the range indicates a real spread in the observed slopes), while in RLQs the slope is flatter (~ 0.5 ± 0.5) and the relative normalisation about × 3 higher (Wilkes & Elvis 1987, Reeves & Turner 2000). The emission mechanisms are different: comptonisation of EUV photons in the Big Blue Bump for RQQs (Gondek et al. 1996) and synchrotron self-Compton scattering of the radio photons in RLQs. In lower luminosity AGN reflected and/or scattered emission from cold/hot material surrounding the X-ray source, such as a corona around the AD or the inner edge of the dusty torus/disk (Mushotzky et al. 1993, Turner et al. 1997, Nandra et al. 1997, Pounds et al. 2001), often dominate the underlying power law. Strong Fe K emission, originating in cold and/or hot material, is present in many low luminosity AGN but weaker/absent at higher luminosities (Reeves & Turner 2000). Please see Brandt (this volume) for a more detailed review of the X-ray emission of AGN.
The most notable difference between RLQs and RQQs is in the radio waveband. In RQQs the SED turns over sharply in the far-IR/mm and radio emission is ~ 100 - 1000 × weaker than in RLQs. In RLQs the IR-radio continuum is smooth with non-thermal emission contributing in both wavebands (Figure 2).
The far-IR cut-off in RQQs is well-determined in only a small number of nearby sources which are bright in the far-IR. Constraints on its slope are frequently steeper than the 2.5 characteristic of homogeneous synchrotron self-absorption. Instead the far-IR emission is identified as grey-body emission from cool dust (Chini et al. 1989, Hughes et al. 1993).
By contrast, the far-IR emission from core-dominated RLQs smoothly extends into the radio, implying a significant/dominant non-thermal component in both wavebands (Figure 2). 3C273 exhibits the correlated variability characteristic of blazars (core-dominated RLQs viewed pole-on), but even here the lack of variability in the hottest part of the IR continuum indicates the presence of hot dust (Tuerler et al. 1999). ESA's Infrared Space Observatory (ISO), the most sensitive IR satellite to date (prior to SIRTF, launched earlier this year), facilitated observation of a larger number of quasars and AGN than in the past. Comparison of the IR continua of RLQs and RQQs suggests that non-thermal IR emission dominates pole-on RLQs but decreases in strength as the viewing angle increases so that thermal emission also contributes in lobe-dominates RLQs (Haas et al. 1998, Polletta et al. 2000).
While both sources in Figure 1 are typical of the SEDs of broad-lined quasars in their class, the relative strengths of the various components range by ~ an order-of-magnitude from source to source, even within the X-ray-bright subset of the population (Elvis et al. 1994). This variety increases with less biased selection techniques such as the X-ray (Kurasziewicz et al. 2003).
In the past 9 years since Figure 1 was published many more SEDs, mainly of low-redshift AGN, have been observed. SEDs have many more data points, particularly in the IR and submm region where ISO, IRAM and the JCMT have made significant steps forward. However the general picture of the SEDs of "normal" AGN has not changed significantly. What has changed much more is our concept of what IS an AGN.