ar X iv : a st ro - p h / 97 11 34 5 v 1 2 8 N ov 1 99 7 COMPTEL OBSERVATIONS OF 3 C 279 DURING THE FIRST 4 YEARS OF THE CGRO - MISSION

The COMPTEL experiment aboard the Compton Gamma-Ray Observatory (CGRO) has observed the gamma-ray blazar 3C 279 several times between April 1991 and September 1995. This paper reports on a consistent analysis of these observations using the most recent COMPTEL data analysis tools. Detections and non-detections of 3C 279 along the CGRO-mission indicate a time-variable MeVflux. Spectral variability is indicated as well, however can not be significantly proven by spectral fitting. The average MeV-spectrum of 3C 279, as measured by COMPTEL over the four-year period, is consistent with a photon power-law slope of ∼ -1.9. This spectrum smoothly connects to the simultaneous 30 MeV to 10 GeV spectrum obtained from an analysis of combined EGRET observations. No spectral break is required to fit the MeVto GeV-spectrum of 3C 279. INTRODUCTION Shortly after the launch of the Compton Gamma-Ray Observatory (CGRO), the blazar-type quasar 3C 279 was detected as an emitter of γ-rays (mainly above 100 MeV) by the EGRET experiment aboard CGRO (Kniffen et al. 1993). This detection stimulated a search for 3C 279 in the contemporaneous COMPTEL data (for an instrument decription see Schönfelder et al. 1993) between 0.75 and 30 MeV, which led to the discovery of 3C 279 at MeV-energies as well (Hermsen et al. 1993, Williams et al. 1995). 3C 279 was redected by COMPTEL roughly two years later (Collmar et al. 1995). Because only parts of the COMPTEL Virgo observations had been analysed in detail, we have started to consistently analyse all of them with emphasis on 3C 279 by applying the most recent COMPTEL analysis techniques and tools. The main objective is to study the MeV-properties of this blazar systematically, and compare them to results in neighbouring energy bands (mainly EGRET). It is known that the MeV-band is an interesting region for blazar spectra (e.g. maximum energy release, spectral breaks) from which insights in the physical emission mechanisms might be derived. In this paper we shall report first results of these analyses covering the time period between April ’91 and September ’95, which corresponds to the Phases I, II, III, and Cycle IV of the CGRO observations. OBSERVATIONS Between April ’91 and September ’95 COMPTEL was pointed towards the Virgo region of the sky several times. All observations for which 3C 279 was within 25oof the pointing direction have been included in our analysis resulting in a total (uncorrected) observation time of 136 days, which converts to an effective observation time (including efficiency corrections and data losses) of ∼40 days. Table 1 gives the relevant dates of the different CGRO viewing periods (VPs). DATA ANALYSIS We have applied the standard COMPTEL maximum-likelihood analysis method (e.g. de Boer et al. 1992) to derive detection significances, fluxes, and flux errors of γ-ray sources in the four standard COMPTEL energy bands (0.75-1 MeV, 1-3 MeV, 3-10 MeV, 10-30 MeV), and a background modelling technique which eliminates any source signature but preserves the general background structure column gives the angular separation between the location of 3C 279 and the pointing direction. Subsequent VPs have been combined in the table. VP Obs. Time Dur. Separ. CGRO Phase # yy/mm/dd yy/mm/dd [days] [] 3 91/06/15 91/06/28 13 8.8 I 11 91/10/03 91/10/17 14 8.5 I 204 206 92/12/22 93/01/12 21 7.2 II 304 306 93/10/19 93/11/09 21 ∼14 III 307 308.6 93/11/09 93/12/01 11 ∼19.5 III 311 313 93/12/13 94/01/03 17 ∼19 III 405 408 94/11/29 95/01/10 32 ∼7 IV (Bloemen et al. 1994). To derive source fluxes, we fitted 3C 279 simultaneously with further known γ-ray sources of the Virgo region (e.g. 3C 273) in an iterative procedure. This iterative approach leads to a simultaneous determination of the fluxes of several potential sources and a background model which takes into account the presence of sources. RESULTS COMPTEL has significantly (≥3σ ) detected 3C 279 at the beginning of its mission (Figure 1) mainly at energies above 3 MeV. The source was not detected during the 3 observational weeks on Virgo (VPs 204 206) in CGRO Phase II, but was redetected in Phase III (Figure 1). In contrast to Phase I, a significant detection is derived in the 1-3 MeV band only. Roughly one year later in CGRO Phase IV, 3C 279 was again invisible for COMPTEL during an observation of 4.5 weeks.