Next Contents Previous

3.4.4. The high energy component

Early gamma ray detections were due to the Cos-B satellite. The source was identified in the data from the position coincidence of a compact source (i.e. a source not resolved by Cos-B) with the position of the quasar [Swanenburg et al. 1978], [Bignami et al. 1981].

The flux above a few MeV is well described by a power law of index 1.4 ± 0.1 [Lichti et al. 1995]. This index has been shown to vary between 1.2 ± 0.2 and 2.2 ± 0.5 [von Montigny et al. 1997], hardening with increasing flux (we give here energy spectral indices rather than photon indices to remain consequent with the discussion of the spectrum at lower energies). Quite expectedly, the spectral index around 1MeV is between the X-ray spectral index and the one observed at higher energies as it is in this region that the spectrum steepens from a slope of about 0.5 to to one of 1.5. It must be stressed that the high energy component described here is a power law and not an exponential cut-off of the medium energy component. This spectral break is an important constraint to any model, it is larger than 0.5, the value expected from simple one component Compton cooling models. Models for the high gamma-ray emission of beamed AGN include the relativistic electron positron beam model of [Marcowith et al. 1995]. In this model the emerging emission is due to inverse Compton process of relativistic electron position pairs on the soft photons from the accretion disc. The observed spectral break is due to the energy dependence of the gamma-ray photosphere defined by the optical depth to pair production being equal to one. A further model is suggested by [Mannheim 1994] and [Mannheim 1993] in which ultra relativistic protons generate gamma ray photons via pion and pair photo production.

Several candidate models have been fitted to the data by [Lichti et al. 1995] and [von Montigny et al. 1997]. They are all based on the assumption that the gamma ray component is emitted by electrons or/and hadrons in the relativistic jet. This assumption is due to the remark that the high energy photon density estimated from the observed flux and the time scale of variability implies that the electron-positron pair production optical depth is considerably larger than one. This would imply that the high energy photons cannot escape from the source region. This is formally described by the dimensionless compactness parameter l:

Equation 2

where sigma is the pair production cross section, L the luminosity and R the size of the source as deduced from the source variability timescale. Using sigma appeq sigmaThompson, a variability time scale of 0.5 days as given by the Ariel V measurement described below and a maximum observed gamma ray luminosity, [Lichti et al. 1995] deduce a compactness of 210, implying that the optical depth of the region to pair production (= l / 4pi) is much larger than 1. Using a more established variability time scale for the medium energy component of several days does relieve the compactness question and lessen the justification for identifying the high energy emission of 3C 273 with the relativistic jet. It should be noted, however, that in other sources, the BL Lacs observed at very high energy, the compactness is such that the gamma ray emission must be emitted by strongly relativistic jets, justifying a similar assumption also in the case of 3C 273.

Next Contents Previous