9512.net
甜梦文库
当前位置:首页 >> >>

XMM-Newton observations of a sample of gamma-ray loud active galactic nuclei


Astronomy & Astrophysics manuscript no. 4921 February 5, 2008

c ESO 2008

XMM–Newton observations of a sample of γ?ray loud active

arXiv:astro-ph/0603268v2 4 Apr 2006

galactic nuclei?
L. Foschini1 , G. Ghisellini2 , C.M. Raiteri3 , F. Tavecchio2 , M. Villata3 , L. Maraschi2 , E. Pian4 , G. Tagliaferri2 , G. Di Cocco1 , G. Malaguti1
1 2 3 4

INAF/IASF-Bologna, Via Gobetti 101, 40129, Bologna (Italy) INAF, Osservatorio Astronomico di Brera, Via Bianchi 46, 23807, Merate (Italy) INAF, Osservatorio Astronomico di Torino, Via Osservatorio 20, 10025, Pino Torinese (Italy) INAF, Osservatorio Astronomico di Trieste, Via G.B. Tiepolo 11, 34131, Trieste (Italy)

Received 27 January 2006; Accepted 9 March 2006
ABSTRACT

Aims. To understand the nature of γ?ray loud active galactic nuclei (AGN) and the mechanisms for the generation of high-energy γ?rays. Methods. We performed a homogeneous and systematic analysis of simultaneous X-ray and optical/UV properties of a group of 15 γ?ray loud AGN, using observations performed with XMM-Newton. The sample is composed of 13 blazars (6 BL Lac and 7 Flat-Spectrum Radio Quasar) and 2 radio galaxies that are associated with detections at energies > 100 MeV. The data for 7 of them are analyzed here for the ?rst time, including the ?rst X-ray observation of PKS 1406 ? 706. The spectral characteristics of the sources in the present sample were compared with those in previous catalogs of blazars and other AGN, to search for di?erence or long term changes. Results. All the selected sources appear to follow the classic “blazar sequence” and the spectral energy distributions (SED) built with the present X-ray and optical/UV data and completed with historical data, con?rm the ?ndings of previous studies on this type of source. Some sources display interesting features: four of them, namely AO 0235 + 164, PKS 1127 ? 145, S5 0836 + 710 and PKS 1830 ? 211 show the presence of an intervening absorption system along the line of sight, but only the last is known to be gravitationally lensed. AO 0235 + 164 was detected during an outburst and its SED shows a clear shift of the synchrotron peak. 3C 273 shows a change in state with respect to the previous BeppoSAX observations that can be interpreted as an increase of the Seyfert-like component and a corresponding decline of the jet emission. This is consistent with the monitoring at radio wavelengths performed during the same period. PKS 1406 ? 706 is detected with a ?ux higher than in the past, but with a corresponding low optical ?ux. Although it is classi?ed as FSRQ, the SED can be modelled with a simple synchrotron self-Compton model.

Key words. Galaxies: active – BL Lacertae objects: general – Quasars: general – X-rays: galaxies

1. Introduction
There is general consensus on the supermassive black hole (SMBH) paradigm as the central engine of active galactic nuclei (AGN). It is more di?cult to obtain quantitative understanding of the physical mechanisms responsible for the observed properties of these cosmic sources. In the AGN zoo, γ?ray loud objects – where with this term we consider the AGN detected at E > 100 MeV – represent a small, but interesting class. Their discovery dates back to the start of γ?ray astronomy, when the European satellite COS-B (1975 ? 1982) detected photons in the 50 ? 500 MeV range from 3C273 (Swanenburg et al. 1978). However, 3C273 remained the only AGN detected by COS-B. A breakthrough in this research ?eld came later with the Energetic Gamma Ray Experiment Telescope (EGRET) on board the Compton Gamma-Ray Observatory (CGRO, 1991-2000). The third catalog of point sources contains 271 sources detected at energies greater than 100 MeV: 93 of them are identi?ed with blazars (66 at high con?dence and 27 at low con?dence), and 1 with the nearby radiogalaxy Centaurus A (Hartman et al. 1999). Among the remaining sources, there are 5 pulsars, the Large Magellanic Cloud, one exceptional solar ?are, and 170 are unidenti?ed. Therefore, EGRET discovered that the blazar-type AGN are the primary source of the extragalactic background in the MeV-GeV range, as suggested by several authors (e.g. Strong et al. 2004, Giommi et al. 2006).
Send o?print requests to: foschini@iasfbo.inaf.it (L. Foschini). ? Based on public observations obtained with XMM–Newton, an ESA science mission with instruments and contributions directly funded by ESA Member States and the USA (NASA).

2

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

Table 1. Main characteristics of the observed AGN. Columns: (1) Name of the source from the Third EGRET Catalog; (2) Name of the known counterpart; (3) Other name; (4) classi?cation of the active nucleus (LBL: low frequency peaked BL Lacertae Object; HBL: high frequency peaked BL Lacertae Object; FSRQ: ?at-spectrum radio quasar; RG: radio galaxy); (5) Coordinates (J2000); (6) Redshift; (7) Galactic absorption column density [1020 cm?2 ] from Dickey & Lockman (1990).
3EG (1) J0222 + 4253 J0237 + 1635 J0530 ? 3626 J0721 + 7120 J0845 + 7049 J1104 + 3809 J1134 ? 1530 J1222 + 2841 J1229 + 0210 J1324 ? 4314 J1339 ? 1419 J1409 ? 0745 J1621 + 8203 J1832 ? 2110 J2158 ? 3023
? ??

Counterpart (2) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714 S5 0836 + 710 Mkn 421 PKS 1127 ? 145 ON 231 3C 273 Cen A PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211 PKS 2155 ? 304

Other Name (3) 3C 66A

4C 71.07

W Comae NGC 5128

AGN Type (4) LBL LBL FSRQ LBL FSRQ HBL FSRQ LBL FSRQ RG FSRQ FSRQ RG FSRQ HBL

α,δ (5) 02 : 22 : 39.6,+43 : 02 : 08 02 : 38 : 38.9,+16 : 36 : 59 05 : 22 : 58.0,?36 : 27 : 31 07 : 21 : 53.4,+71 : 20 : 36 08 : 41 : 24.3,+70 : 53 : 42 11 : 04 : 27.3,+38 : 12 : 32 11 : 30 : 07.0,?14 : 49 : 27 12 : 21 : 31.7,+28 : 13 : 59 12 : 29 : 06.7,+02 : 03 : 09 13 : 25 : 27.6,?43 : 01 : 09 13 : 37 : 39.8,?12 : 57 : 25 14 : 08 : 56.5,?07 : 52 : 27 16 : 32 : 32.0,+82 : 32 : 16 18 : 33 : 39.9,?21 : 03 : 40 21 : 58 : 52.0,?30 : 13 : 32

z (6) 0.444 0.94 0.05534 > 0.5? 2.172 0.03002 1.184 0.102 0.15834 0.00182?? 0.539 1.494 0.0247 2.507 0.116

NH (7) 8.99 8.95 3.33 3.81 2.91 1.38 4.02 1.88 1.79 8.62 4.82 2.77 5.47 21.9 1.69

Lower limit evaluated on the basis of the non-detection of the host galaxy (Sbarufatti et al. 2005). This redshift is not indicative and the distance of 3.84 Mpc is adopted here. See Evans et al. (2004) for more details.

During the years following these discoveries, much e?ort has been dedicated to the identi?cation of the remaining 170 sources. This is a challenging enterprise given the large position error contours of EGRET sources (typically ? 0? .5 ? 1? ). A signi?cant advancement has been obtained by Sowards-Emmerd et al. (2003, 2004), who performed a radio survey at 8.4 GHz. By using a “?gure of merit” obtained combining the 8.4 GHz ?ux, the radio spectral index and the X-ray ?ux (when available), they proposed 20 new identi?cations of EGRET sources with blazar-type AGN. A strong improvement in the identi?cation and discovery of new γ?ray loud AGN is expected with the forthcoming missions GLAST 1 and AGILE2 . An important complement to these discoveries was the observations performed by the Italian-Dutch satellite BeppoSAX (1996-2002) and operating in the 0.1 ? 300 keV energy band (see Ghisellini 2004 for a review on blazar observations with this satellite). During its lifetime, BeppoSAX observed more than 80 blazars (Giommi et al. 2002, Donato et al. 2005) and sampled with high sensitivity the X-ray region of the spectral energy distribution (SED) over more than three decades in energy. The observations of these and other high-energy satellites, together with ground telescopes, led to the discovery that the spectral energy distribution (SED) of blazars is typically composed of two peaks, one due to synchrotron emission and the other to inverse Compton radiation, the latter discovered by CGRO/EGRET (von Montigny et al. 1995). Maraschi et al. (1995b) and Sambruna et al. (1996) noted that the broad-band spectra of BL Lac and Flat-Spectrum Radio Quasars (FSRQ) share common features and properties (that justi?ed the common designation of “blazars” proposed by Spiegel in 1978). Fossati et al. (1998) and Ghisellini et al. (1998) proposed a uni?ed scheme where the blazars are inserted into a “sequence” according to their physical characteristics. Low luminosity BL Lac have the synchrotron peak in the UV-soft X-ray energy band and therefore are “highenergy peaked” (HBL). As the synchrotron peak shifts to low energies (near infrared, “low-energy peaked” BL Lac or LBL), the luminosity increases and the X-ray emission can be due to synchrotron or inverse Compton or a mixture of both. In the case of FSRQ, the blazars with the highest luminosity, the synchrotron peak is in the far infrared and the X-ray emission is due to inverse Compton radiation. However, other authors reported failures in the above mentioned scheme (see, for example, Padovani et al. 2003, Landt et al. 2006). The two-peak SED is a dynamic picture of the blazar behaviour: indeed, these AGN are characterized by strong ?ares during which the SED can change dramatically (e.g. Tagliaferri et al. 2002). The most striking example of such a behaviour is represented by Mkn 501 – although at γ?ray energies this source was not detected by EGRET, but by TeV telescopes – that during an outburst in 1997 showed a shift of the synchrotron peak to the hard X-ray energy band (50 ? 100 keV, Pian et al. 1998). The variability on di?erent time scales and, particularly, the intraday variability, is one of the striking characteristics of blazars and is considered one of the proofs that the continuum is generated by a relativistic jet with a small observing angle (for a review see Wagner & Witzel 1995, Ulrich et al. 1997).
1 2

http://www-glast.stanford.edu http://agile.iasf-milano.inaf.it

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

3

XMM-Newton (launched in Dec. 1999, Jansen et al. 2001) is a satellite with the current largest collecting area, useful for timing studies, together with a good sensitivity and spectral resolution. XMM-Newton covers a lower frequency range than BeppoSAX, i.e. from the optical/UV domain to the X-rays, up to 10 keV. From the large XMM-Newton public archive, we have selected and analyzed all the publicly available EGRET-detected AGN data. Although the single observations were originally intended for other purposes, it was possible to carry out a homogeneous analysis and to study the main spectral characteristics of these sources in the X-ray energy band. Several sources of the present sample deserve particular attention and many detailed studies have been published on the individual sources. However, the aim of the present work is to perform an overall view and comparison with previous surveys to search for common features that could explain the γ?ray generation. Some early resuls of the present work have been presented in Foschini et al. (2006). This paper is organized as follows: in Sect. 2 the selection criteria, biases and the parameters used in the data analysis are presented; in Sect. 3, the spectral characteristics are shown and compared with other catalogs. The absorption systems are discussed in Sect. 4, and a short note on the X-ray spectral features is presented in Sect. 5. The spectral energy distributions and the blazar sequence is shown in Sect. 6. Sect. 7 contains ?nal remarks. In Appendix A we provide the notes on the individual sources together with the ?ts and some tabular material. The cosmology values adopted through the paper, when not explicitly declared, are H0 = 70 km·s?1 Mpc?1 , ?λ = 0.7 and ?m = 0.3.

2. Sample selection and data analysis
The starting sample consists of all the AGN in the Third EGRET Catalog (Hartman et al. 1999) updated with the recent results by Sowards-Emmerd et al. (2003, 2004). This sample has been cross-correlated with the public observations available in the XMMNewton Science Archive3 to search for spatial coincidences in the ?eld of view (FOV) of the EPIC camera, within 10′ of the boresight4. 15 AGN have been found (Table 1) as of January 4th , 2006, and for three of them there are more than 5 observations available (see the observation log in Table A.1), making it possible also to study the long term behaviour.

2.1. Biases and caveats
The present work su?ers from several biases, but nevertheless it is possible to obtain useful information about the overall behaviour of the γ?ray loud AGN. The ?rst source of bias is the Third EGRET catalog itself: the large point spread function (PSF) of the EGRET telescope and its moderate sensitivity, changes in the position from the 2EG to the 3EG catalog, double (or more?) sources not resolved by the EGRET PSF (see the notes on the single sources). Sowards-Emmerd et al. (2005) called for the release of a Fourth EGRET catalog, but a major advancement will be possible when the GLAST satellite is operational. The arcminute-sized PSF of the LAT telescope and the higher sensitivity would then improve the con?dence of the suggested associations (e.g. 3EG J0530 ? 3626 or 3EG J1621 + 8203) or disentangle the multiple contributions of certain EGRET sources (see, for example, 3EG J0222 + 4253). The use of the XMM-Newton public data introduces new biases. Exposures and instrument modes were not selected for a survey, but with completely di?erent purposes (e.g. calibration, ToO, ...). In the XMM-Newton archive there are many more pointings than the 46 reported here: some were discarded because of problems in the processing of the observation data ?les (ODF), some others were not used because the instrument mode does not match the major part of the pointing of the same source5 , or because the observations were still covered by the PI proprietary data rights at the time when the archive was scanned. In spite of 46 observations, the present sample is made of only 15 AGN. Three sources dominate: 3C 273 with 15 observations, PKS 2155 ? 304 with 9, and Mkn 421 with 6 (see the observation log in Table A.1). The large di?erences in settings of the instrument modes between the individual observations had a particularly severe impact on the OM data (Table A.7). It is not possible to have one single ?lter to be used as a reference for all the observations. In the best case, the magnitudes with ?lter UVW1 are available for 18 of 46 pointings. In one case only (PKS 1830 ? 211), the optical counterpart of the blazar has a V magnitude ≈ 25 (Courbin et al. 2002) and therefore is beyond the instrument capabilities.

2.2. Common procedures of data analysis
For the processing, screening, and analysis of the data from the EPIC MOS1, MOS2 (Turner et al. 2001) and PN cameras (Str¨ der u et al. 2001), standard tools have been used (XMM SAS v. 6.1.0 and HEAsoft v 6.0). The standard procedures described in Snowden et al. (2004) were followed. Only single pixel events have been selected, excluding border pixels or columns with higher
http://xmm.vilspa.esa.es/external/xmm data acc/xsa/index.shtml This maximum distance has been selected to take into account that within that region the telescope vignetting is well corrected according to Kirsch (2005). 5 This is the case of Mkn 421, 3C 273, and PKS 2155-304, that are calibration sources. For the purposes of the present work, only the pointings with the PN detector in small window mode were used.
4 3

4

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

Table 2. Best ?t model parameters for the 15 AGN (present work). In the case of sources with multiple pointings, the weighted averages are shown. Since PKS 2155 ? 304 (3EG J2158 ? 3023) is best ?tted in 4 pointings with the broken power law model and in the remaining 5 with the simple power law model; we reported the averages of both models. Columns: (1) Name of the source; (2) absorbing column density [1020 cm?2 ]; (3) photon index Γ, if the best ?t is a simple power law model or soft photon index Γ1 , if the best ?t model is a broken power law; (4) hard photon index Γ2 for the broken power law model; (5) Break energy [keV].
Name (1) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714 S5 0836 + 710 Mkn 421 PKS 1127 ? 145 ON 231 3C 273 Cen A PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211 PKS 2155 ? 304 NH (2) Gal. Gal. Gal. Gal. 14 ± 3 Gal. 12+2 ?1 2.5 ± 0.6 Gal. 1523 ± 261 6.7 ± 0.9 Gal. 14 ± 1 63 ± 1 1.69 ± 0.06 Gal. Γ/Γ1 (3) +0.12 2.91?0.08 2.33 ± 0.04 1.95 ± 0.03 2.70 ± 0.02 1.379 ± 0.007 2.38 ± 0.09 +0.08 1.40?0.05 2.77 ± 0.04 2.02 ± 0.08 2.22 ± 0.06 1.80 ± 0.04 1.59 ± 0.01 +0.08 2.11?0.06 1.00 ± 0.09 2.9 ± 0.1 2.7 ± 0.1 Γ2 (4) +0.10 2.23?0.09 2.1 ± 0.1 1.74 ± 0.03 +0.08 1.98?0.09 ? 2.7 ± 0.2 1.22 ± 0.06 ? 1.67 ± 0.05 ? ? ? 1.78 ± 0.07 1.32 ± 0.06 ? 2.94 ± 0.06 Ebreak (5) 1.3 ± 0.2 +0.7 3.3?0.5 +0.3 1.5?0.2 +0.2 2.3?0.1 ? 2.7 ± 1.0 +1.0 2.7?0.8 ? 1.44 ± 0.08 ? ? ? +0.3 2.5?0.4 3.5 ± 0.7 ? 2.7 ± 0.7

Table 3. Best ?t model parameters for the 15 AGN from the BeppoSAX catalogs by Giommi et al. (2002) and Donato et al. (2005). The values are the weighted averages obtained from multiple observations (if any) and from the two catalogs. In the case of 3C 273 (3EG J1229 ? 0210) there were 9 observations, 5 best ?tted with a broken power law and 4 with a single power law model: both model averages are reported. For NGC 6251 (3EG J1621 + 8203) the average is obtained from Chiaberge et al. (2003) and Guainazzi et al. (2003) from a ?t in the energy band 0.1 ? 200 keV. Grandi et al. (2003) is the reference for Cen A (3EG J1324 ? 4314), from a ?t in the energy band 0.4 ? 250 keV excluding the 5.5 ? 7.5 keV band (iron emission line). The value for PKS 1830 ? 211 (3EG J1832 ? 2110) has been obtained by De Rosa et al. (2005) from a ?t in the energy band 0.5 ? 80 keV (Chandra and INTEGRAL). Columns: (1) Name of the source; (2) absorbing column density [1020 cm?2 ]; (3) photon index Γ, if the best ?t is a simple power law model or soft photon index Γ1 , if the best ?t model is a broken power law; (4) hard photon index Γ2 for the broken power law model; (5) Break energy [keV].
Name (1) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714 S5 0836 + 710 Mkn 421 PKS 1127 ? 145 ON 231 3C 273 Cen A PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211 PKS 2155 ? 304 NH (2) Gal. Gal. Gal. Gal. 78+55 ?35 Gal. Gal. Gal. Gal. Gal. 1020+90 ?40 ? ? 9±1 194+28 ?25 Gal. Γ/Γ1 (3) 2.22 ± 0.06 2.0 ± 0.1 1.74 ± 0.02 2.5 ± 0.2 1.31 ± 0.02 1.9 ± 0.2 1.42 ± 0.05 2.58 ± 0.01 2.0 ± 0.1 1.58 ± 0.03 +0.03 1.80?0.04 ? ? 1.79 ± 0.06 1.09 ± 0.05 2.3 ± 0.1 Γ2 (4) ? ? ? 1.8 ± 0.1 ? 2.3 ± 0.3 ? 1.52 ± 0.06 1.603 ± 0.006 ? ? ? ? ? ? 2.8 ± 0.1 Ebreak (5) ? ? ? 3.0 ± 0.4 ? 1.3 ± 0.8 ? 2.8 ± 0.2 0.9 ± 0.3 ? ? ? ? ? ? 1.7 ± 0.2

o?set. High-background ?ares a?ected the observations randomly, and in some cases it was necessary to ?lter the available data. Time intervals contaminated by ?ares have been excluded by extracting the whole detector lightcurve for E > 10 keV and by removing the periods with count rates higher than 1.0 s?1 for PN and 0.35 s?1 for MOS, as suggested by Kirsh (2005).

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

5

The source spectra have been extracted from a circular region with a radius of 40′′ and centered in the catalog (radio) position of the AGN. The background to be subtracted in the analysis was derived from a circular region, with the same radius, near the selected source. In the case of pile-up, the source region is an annulus with still an external radius of 40′′ and an internal radius selected to minimize the pile-up e?ects by using the task epatplot of XMM SAS. It resulted 8′′ for 3C 273 and PKS 2155 ? 304, and 10′′ for Mkn 421. For Cen A, the annulus has internal and external radii of size 20′′ and 50′′ , respectively. Since PN is the most stable detector, with negligible degradation of performance to date, we adopted it as the prime instrument. Data from MOS cameras have been used only when it was necessary to check a ?nding obtained with the PN detector or to increase the statistics of a speci?c observation. For very bright sources (with ?uxes of the order of 10?10 erg cm?2 s?1 or more), only the PN data in small window mode have been analyzed. The spectra were rebinned so that each energy bin contained a minimum of 20 counts and ?t in the 0.4 ? 10 keV energy range for the PN detector and 0.5 ? 10 keV for MOS detectors, because of the uncertainties in the calibration and cross-calibration at lower energies (cf. Kirsch 2005). The ?uxes and luminosities were calculated in the 0.4 ? 10 keV band by extrapolating the model spectrum with the command extend of xspec. The photon redistribution matrix and the related ancillary ?les were created appropriately with the rmfgen and arfgen tasks of XMM-SAS. The data from the Optical Monitor (Mason et al. 2001) were also reprocessed with the latest version of SAS. Through the paper, we report only the ?ts with reduced χ2 less than 2 (χ2 < 2) and we consider signi?cant improvement ? in the ?t with Ftest > 99%. In the case of an added spectral component, we evaluated the improvement of the ?t by the ?χ2 value (cf Protassov et al. 2002). All the quoted uncertainties in the parameters are at the 90% con?dence level for 1 parameter (?χ2 = 2.71), unless otherwise stated.

3. Average spectra and comparisons with other catalogs
Most of the sources analyzed here are blazars, except for two radiogalaxies Fanaro?-Riley Type I (FRI), that are thought to be blazar-like sources seen at large viewing angles (Urry & Padovani 1995). Blazars display featureless X-ray continuum, occasionally with hints of breaks, curvature, or, more seldom, a soft excess (e.g. Giommi et al. 2002, Donato et al. 2005, Perlman et al. 2005). Radiogalaxies can have a much more complex environment at low energies (e.g. Evans et al. 2005, Grandi et al. 2005). However, since the main purpose of this work is to study and compare the continuum properties, we decided to ?t the X-ray spectra with a redshifted power-law model (zpo in xspec; see Table A.2) and a broken power law model (bknpo in xspec; see Table A.3). The absorption can be ?xed to the Galactic column density (Dickey & Lockman 1990) or ?tted. Sometimes a more complex ?t is necessary and is analyzed separately. More details on the ?ts are available in Appendix A (tables and notes on the individual sources). Several of the 15 sources analyzed here show signi?cant variability, particularly on small time scales. Therefore, the comparison of average spectral parameters presented here (Table 2) and the values available in the BeppoSAX catalogues by Giommi et al. (2002) and Donato et al. (2005), can be considered a indicator of long term spectral variability, since XMM-Newton observations refer to the period 2000 ? 2004 and BeppoSAX data have been collected in the period 1996 ? 2002. The spectra of the present work are generally best ?tted with broken power law models (9/14 in Table 2 without PKS 2155 ? 304, that is ?tted with both models), compared to only 4/14 sources in BeppoSAX data (Table 3, without 3C 273, that is ?tted with both models). There is a possible important bias factor in BeppoSAX ?ts. The Italian-Dutch satellite concentrators LECS and MECS have energy bands overlapping at ≈ 2 keV, where most of the blazars have the break energy. Therefore, it is also possible that – in some cases – the intercalibration constant between the two detectors could have “absorbed” some spectral shape variations, thus leading to prefer the single power law model. This could be the case of PKS 1127 ? 145, that is known to have an intervening system along the line of sight, but the BeppoSAX ?t does not require an additional absorption. In other cases, there are changes in the source state (e.g. AO 0235 +164, that in the present observation was found in outburst) or 3C 273, as already noted by Page et al. (2004). In the latter case, the blazar shows in the XMM-Newton data an increase of the break energy and a small softening of Γ2 , the photon index at E > Ebreak . Also Cen A shows a softer photon index and a ?ux lower by a factor of 4 with respect to the BeppoSAX data analyzed in Grandi et al. (2003). In the two HBL (Mkn 421 and PKS 2155 ? 304), the broken power law model appears to be the simpli?cation of a more complex curved model, as already noted by Brinkmann et al. (2001, 2003), Sembay et al. (2002), Ravasio et al. (2004). Comparing the parameters in Table 2 with the larger catalogs by Giommi et al. (2002), Donato et al. (2005), Evans et al. (2005), Grandi et al. (2005) containing also radio-loud AGN not detected by EGRET, there are no signs of di?erences between γ?ray loud and quiet AGN. The analysis of the three most intensively observed sources is in agreement with the typical behaviour of these types of sources. Even though they require a broken power law model, to have an overall view of the source behaviour, we correlate the photon index of the single power law model with its normalization at 1 keV. This gives an idea of the behaviour of the spectral shape with ?ux variations. PKS 2155 ? 304 shows a clear correlation (linear correlation coe?cient r = ?0.76 for 8 observations and p ≤ 0.035), with a hardening of the photon index with increasing ?ux (Fig. 1, bottom right) if one point is not considered (ObsID 0124930301). While the behaviour of most points can be explained by considering a ?ux increase with a constant synchrotron peak (implying

6

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

Fig. 1. Correlations: Photon index versus normalization for (top left) 3C 273 (3EG J1229 + 0210), (bottom left) Mkn 421 (3EG J1104 + 3809), and (bottom right) PKS 2155 ? 304 (3EG J2158 ? 3023). (top right) Equivalent width of the FeKα line versus photon index for 3C273. The open square indicates the XMM-Newton data from the present work, while the ?lled circle represents the values from BeppoSAX observations reported in Grandi & Palumbo (2004). The points of the simultaneous BeppoSAX-XMM-Newton observation (June 2001) are tagged with a triangle, ?lled for BeppoSAX and open for XMM-Newton.

a hardening of the spectrum), the outlying point can be explained as due to a frequency shift of the synchrotron peak. For this ObsID, OM data (Table A.7) suggest a high ?ux level in the bands U, B, and V, but, again, available magnitudes are fragmentary and it is not possible to search for more stringent correlations. The correlation in Mkn 421 is poor (r = ?0.60 for 6 points, p ≤ 0.2), although a general trend of spectral hardening with ?ux increasing can be noted (Fig. 1, bottom left). 3C 273 shows instead a spectral softening with ?ux increasing (Fig. 1, top left, r = 0.66 with 15 observations, p ≤ 0.0077), although there is a non negligible scatter of the points, suggesting that other processes are playing important roles in this sources. Indeed, it is known that this blazar also has a Seyfert-like component that can be detected (Grandi & Palumbo 2004). Fig. 1 (top right) shows the equivalent width of the FeKα line versus the photon index for both BeppoSAX observations (?lled circles; data from Grandi & Palumbo 2004) and the present work (open squares; for XMM-Newton see Table A.5). 3C 273 was in a di?erent state during the two satellites observations with only one point overlapping, when using the simultaneous BeppoSAX and XMM-Newton observation performed in June 2001 (ObsID 0136550101, triangles in Fig. 1 top right) to cross-calibrate the respective instruments (cf Molendi & Sembay 2003). This strenghtens the validity of the other values obtained by the two satellites as indicating an e?ective change in the state of the source. The general trend can be understood, in the framework of the Grandi & Palumbo (2004) results on 3C 273 and the more general picture on blazars outlined by Maraschi & Tavecchio (2003), as a weakening of the jet component and an increase of the Seyfert-like part. With respect to the BeppoSAX observations (1996?2002), we noted in the XMM-Newton observations (2000?2004), an increase of the “thermal” component and a softening of the hard photon index The former is indicated in the broken power law model by a shift to high energies of the break energy

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

7

(see Table 2: the average value for XMM-Newton is Ebreak = 1.44 ± 0.08 keV compared to the BeppoSAX value of 0.9 ± 0.3 keV); in the blackbody plus power law model (Table A.4) this is indicated by an increase of the temperature (from the average value of 54+6 eV measured by BeppoSAX to 143 ± 6 eV derived from the data analyzed in the present work). This behaviour of the ?4 continuum is accompanied by an increase of the equivalent width of the (broad) iron emission line either at 6.4 or 6.7 keV (Fig. 1, top right). According to the results in Table A.5, the major improvements in the ?t occur with the detection of the (broad) iron line centered at 6.4 or 6.7 keV, but sometimes there is no detection at all. If we consider that the detection and the energy centroid are related to di?erent degrees of ionization (see, for example, the review by Reynolds & Nowak 2003), the interpretation of the data still favours the hypothesis of an increase of the accretion around the SMBH. A more detailed spectral analysis is required to better assess the state variation of 3C 273, but this is outside the scope of the present work. However, the general trend outlined here is also con?rmed by the radio data reported, for example, by Ter¨ sranta et al. (2005) with observations at 22 and 37 GHz: a there is a decreasing of activity from 1996, with an outburst in 2003, but smaller than in 1998 ? 1999. Another interesting case in the present sample is AO 0235+164, that displays the typical characteristics of blazars in outburst. This should be compared with the negative detection of variability reported in the 2004 observations (Raiteri et al. 2005). A more detailed analysis of all the XMM-Newton data sets (both public and private) is available in Raiteri et al. (2006). In the present observation, the source displays a shift of the synchrotron peak (see the discussion in Section 6), with a hint of periodicity of 7829 s (signi?cance 4.5σ), but with a low quality factor, because of a limited number of cycles (2.15). An inspection of the lightcurve suggests that this periodicity is transient: periodic ?ares appear during the activity phase of the source, while X-ray observations of the source in quiescence have detected constant ?ux and, obviously, no periodicities (Raiteri et al., 2006).

4. Intervening absorption
Torres et al. (2003) have suggested that gravitational microlensing can boost the γ?ray ?ux. Therefore, we searched in the present sample for any presence of intervening systems of any type. Damped Lyman?α (DLA) systems (see Wolfe et al. 2005 for a review) are present along the line of sight toward AO 0235+164 and PKS 1127?145. In the case of AO 0235+164 (z = 0.94, Cohen et al. 1987), the intervening system is placed along the line of sight at z = 0.524 (see, e.g., Raiteri et al. 2005 for a more recent discussion on this). This intervening system has been measured z by ROSAT and ASCA obtaining a value of NH = (2.8 ± 0.4) × 1021 cm?2 (Madejski et al., 1996). In addition, Raiteri et al. (2005) z 21 measured a value of NH = (2.4 ± 0.2) × 10 cm?2 with an XMM-Newton observation performed on January 18, 2004. The present spectrum in the 0.4 ?10 keV energy band for PN and the 0.5 ?10 keV band for the two MOS detectors is the best ?t with a broken z power law model absorbed by the Galactic column together with one at redshift z = 0.524 with NH = (2.56 ± 0.05) × 1021 cm?2 . The low energy photon index is Γ1 = 2.50 ± 0.01 and the high-energy one is Γ2 = 2.05 ± 0.03, with the break at 3.08+0.09 keV, ?0.12 for a χ2 = 1.10 and 801 dof (see also Appendix A). ? PKS 1127 ? 145 is another quasar with an additional absorber at z = 0.321 along the line of sight, probably due to two late-type galaxies (Bergeron & Boiss? 1991, Lane et al. 1998). The best ?t model is the broken power law, with the z = 0.321 e absorption column of (1.2+0.2) × 1021 cm?2 , the Galactic column, and no intrinsic absorption. Chandra observations reported the ?0.1 same absorption, but the continuum is best ?tted with a single power law with Γ = 1.19 ± 0.02 (Bechtold et al. 2001). The absorption in X-rays, in both the above cases, is lower than that measured from optical observations, but it can be due to di?erent metallicity (Z < Z⊙ ) or to the presence of dust with a ratio di?erent from that in the Galaxy, that is common in DLA systems (Pei et al. 1991, Pettini et al. 1994). However, what is important in these systems is the presence of galaxies along the line of sight that can cause gravitational lensing e?ects, which in turn enhance the ?ux of the background blazar. Another blazar (S5 0836+710) presents an intervening system at z = 0.914, but optical observations by Stickel & K¨ hr (1993) u indicated the presence of Mg II λλ2796, 2803, which do not qualify this system as damped Lyman?α. The blazar spectrum, already quite hard, presents a spectral ?attening at low energies. This type of ?attening has been observed in other blazars at high redshift (z > 2; see, e.g., PMN J0525 ? 3343 in Fabian et al. 2001 or on RBS 315 in Piconcelli & Guainazzi 2005) and a few hypotheses have been invoked, like instrinsic absorption or intrinsic spectral properties. The best ?t for S5 0836 + 710 is that with the additional absorption at the redshift of the quasar and no improvements are obtained by adding an absorber at the redshift of the intervening system (z = 0.914). Therefore, it appears that this system is not responsible for the additional absorption or – at least – the present data do not allow us to separate the di?erent components, if any. Radio observations by Hutchison et al. (2001) show that near the blazar core the polarization is very low with respect to typical values in other quasars. The authors explained this by the presence of an external environment surrounding the jet (a cocoon?). Such a plasma cocoon could act like the “warm absorber” of Seyferts and explain the spectral ?attening at low energies. Yet another hypothesis is a low-energy cut-o? in the 2 relativistic electron distribution at γmin , which would yield a ?attening below ≈ γmin Γ2 νext , where νext is the peak frequency of the seed photon radiation ?eld. These hypotheses are discussed in, e.g., Fabian et al. (2001). Two more sources in the present sample show evidence of intrinsic absorption and are the two radiogalaxies (Cen A and NGC 6251). This is known to be due to the environment in the radiogalaxies (cf Evans et al. 2005, Grandi et al. 2005). The last case of intervening systems is PKS 1830 ? 211 (z = 2.507), that is gravitationally lensed by a galaxy at z = 0.886 found by Wiklind & Combes (1996) through infrared observations of hydrocarbon absorption lines. The ?t with a redshifted power law model absorbed by the Galactic column and the intervening system at z = 0.886, gives results consistent with previous

8

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

Fig. 2. SED of newly modelled blazars: (top, left) AO 0235 + 164; (top, right) PKS 1334 ? 127; (bottom, left) PKS 1406 ? 076; (bottom, right) PKS 1830 ? 211, not corrected for the gravitational lensing e?ects. The black line is a single–zone synchrotron inverse Compton model using the input parameters listed in Table 4. In all the SED, the XMM–Newton data are indicated with black ?lled symbols, while the remaining symbols and lines (grey) refer to archival data. The bump at ? 1014?15 Hz is the assumed disk spectrum.

X-ray analyses by Mathur & Nair (1997), Oshima et al. (2001), and De Rosa et al. (2005). The absorption due to the intervening z system averaged over the three observations is NH = (2.3 ± 0.1) × 1022 cm?2 and the photon index is Γ = 1.14 ± 0.02. However, some residuals at low energy are present and an improvement in the ?t (with ?χ2 = 30.9, 103, and 16.6 for a decrease of two degrees of freedom, for the three observations respectively) can be obtained by adding a thermal plasma model (mekal) at the redshift of the blazar, with solar abundances and temperature kT = 0.39 ± 0.06 keV. The absorption of the intervening z system is slightly greater (NH = (3.0 ± 0.4) × 1022 cm?2 ) and the photon index a little steeper (Γ = 1.21 ± 0.04). This thermal plasma can be the “warm absorber” suggested by Fabian et al. (2001) to explain the X-ray de?cit at low energies in high redshift blazars. Interestingly, the statistical best ?t of these three XMM-Newton observations is obtained with a broken power law model and lower absorption (Table A.3), thus suggesting that the low X-ray de?cit could be due to something intrisic to the electron distribution.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

9

Table 4. Parameters for the models of AO 0235 + 164, PKS 1334 ? 127, PKS 1406 ? 076, and PKS 1830 ? 211: above the dividing line are listed the input parameters used to model the SED according to the ?nite injection time synchrotron–inverse Compton model of Ghisellini, Celotti & Costamante (2002). Below the dividing line we list the output parameters for the four ?tted sources. See the text for more details.
Parameter R ?R L′ inj γbreak γmax s B Γ θ δ L BLR R BLR γpeak UB Ur′ (γpeak ) LB Le Lp Lrad AO 0235 + 164 4.5 2.8 7.5 2.0 × 104 6.0 × 105 2.55 1.3 16 3 18.2 3.0 7.0 2.0 × 104 0.067 0.042 3.28 1.05 7.16 4.588 PKS 1334 ? 127 3.5 3.5 3.0 2.5 × 102 4.0 × 103 2.7 3.7 10 6 9.6 4.5 3.5 2.5 × 102 0.54 1.8 6.26 1.78 3.15 3.59 PKS 1406 ? 076 2.0 2.0 3.0 1.0 × 102 5.0 × 103 2.2 1.0 10 3.6 14.3 0.8 5.0 1.6 × 103 0.04 0.25 0.15 8.8 25 3.2 PKS 1830 ? 211 2.0 30 7.0 1.4 × 102 1.6 × 103 2.8 3.8 17 3.5 16.4 12 3.5 1.4 × 102 0.57 10.87 14 1.0 26.5 20.8 Units 1016 cm 1015 cm 1043 erg s?1

Gauss degree 1044 erg s?1 1017 cm erg cm?3 erg cm?3 1045 erg s?1 1045 erg s?1 1046 erg s?1 1045 erg s?1

5. X-ray spectral line features
The spectra of the AGN in the present sample sometimes show some features, but these are generally within 2σ deviations from the best ?t model. For more signi?cative detections (e.g. in PKS 2155 ? 304, known to be due to warm-hot intergalactic medium, see Cagnoni et al. 2004), we bypassed the problem with a proper selection of the energy band, since the study of these features is outside the aims of the present work. No evidence (> 3σ) of features linked to physical characteristic of any cosmic source is found in the present data set, except for the iron line complex of Cen A (see Table A.6 and the note on this source in Appendix A) and some detections in 3C 273 (Table A.5). The latter is not always evident, although a forced ?t with a broad iron line both neutral and ionized can give sometimes a non negligible improvement in the χ2 . The implications have been already discussed in the Sect. 3.

6. Spectral Energy Distributions and the blazar sequence
SED have been constructed and modelled to study the multiwavelength emission over a broad energy range. All but two (PKS 1334 ? 127 and PKS 1830 ? 211) of the blazars in the present sample have been studied in detail by Ghisellini et al. (1998), Tagliaferri et al. (2000), Ghisellini, Celotti & Costamante (2002). The radiogalaxies have been studied in Chiaberge et al. (2001, 2003), Guainazzi et al. (2003), Foschini et al. (2005), and Ghisellini et al. (2005). We refer to these papers and to the references therein. The SED with XMM-Newton data are reported in Fig. A.1 and A.2, together with the model of synchrotron and inverse Compton radiation (including self-Compton and external Compton) in a homogeneous region applied to the data. The results obtained do not change dramatically with respect to the above mentioned works. There are however four cases, namely AO 0235 + 164, PKS 1334 ? 127, PKS 1406 ? 76 and PKS 1830 ? 211, which are worth investigating further. AO 0235 + 164 shows a clear shift in its peak frequency, as expected during ?aring activities of blazars: the present XMM-Newton observation was performed when the source was in outburst (see Sect. 3). PKS 1334 ? 127 has been associated with an EGRET source only in the Third Catalog (Hartman et al. 1999), and therefore is missing in Ghisellini et al. (1998). PKS 1406 ? 076 was never detected in X-rays and PKS 1830 ? 211 is a gravitationally lensed system, also missing in Ghisellini et al. (1998). We therefore applied the same model used in Ghisellini, Celotti & Costamante (2002) in order to ?nd out the physical parameters of the four sources. The main assumptions of the model can be summarized as follows: the geometry of the source is a cylinder – except for PKS 1830 ? 211, that is analyzed later – of radius R and length, in the comoving frame, ?R′ = R/Γ, where Γ is the bulk Lorentz factor; θ is the viewing angle, δ the Doppler factor, and B the magnetic ?eld. The radiating particle distribution is assumed to be N(γ) ∝ γ?s , where the value of s depends on the value of γ of the injected particles. The injected power in the comoving frame is L′ . The external seed photon ?eld has a dimension RBLR and luminosity LBLR . It is assumed inj to mainly originate in a Broad Line Region or any other external source and it is calculated as a fraction of the disk luminosity

10

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

0235

1830-211 0235 1406-076 1334-127

Fig. 3. Updated blazar sequence, adapted from Ghisellini, Celotti & Costamante (2002), with the addition of the new TeV BL Lacs. We mark with black ?lled symbols the sources analyzed in this paper, emphasizing AO 0235 + 164 (in quiescence and during the outburst presented in this work), PKS 1334 ? 127, PKS 1406 ? 076 (the ?t without the X-ray detection and with the present data), and PKS 1830 ? 211.

(generally 10%). The magnetic and radiative energy densities are indicated with U B and Ur , respectively. Electron, proton and radiation powers are represented by Le , Lp , and Lrad , respectively. Fig. 2 show the model with the SED, while in Table 4 we list the input and output parameters for the model. Only AO 0235 + 164 displays signi?cant variability in the model parameters with respect to Ghisellini et al. (1998), but it continues to ful?ll the requirements of the blazar sequence. PKS 1334 ? 127 behaves as a typical FSRQ. The X-ray emission of PKS 1406 ? 076 is modelled as due to the inverse Compton emission from synchrotron seed photons, although this source is a FSRQ. An external source of seed photons is needed to generate γ?rays in the EGRET energy band (not simultaneous data) and there is an apparent anti-correlation between X-ray and optical/UV emission: indeed, this ?rst X-ray detection is simultaneous with low optical ?ux, while archival data (not simultaneous) report higher optical ?ux and only an upper limit in X-rays (for more details, see the note on this source in the Appendix). This does not allow us to claim an anti-correlation. In the case of PKS 1830 ? 211, there are some problems that should be taken into account: the magni?cation e?ects of the gravitational lensing are still uncertain (cf Oshima et al. 2001 and Courbin et al. 2002) and so as the absorption, both due to the Galactic column (the source is apparently located at low Galactic latitude, with Galactic coordinates l = 12? .16 and b = ?5? .71) and to the intervening system. Therefore, instead of making hypotheses about the quantity and quality of corrections to be applied, we decided to analyze the observed SED without any correction. This means that the anomalies in the parameters of PKS 1830 ? 211 (Table 4) re?ect the uncertainties in the magni?cation and the absorption. For example, the needed LBLR necessary for the external Compton contribution was calculated as 30% of the disk luminosity, while for the three other blazars a value of 10% was taken. A proper dereddening could result in the needed optical ?ux, without invoking an increase of percentage of the disk luminosity. The location of the sources analyzed here in the blazar sequence can be seen in Fig. 3, where we show γpeak , the Lorentz ′ factor of the electrons radiating mostly at the peak of the SED, versus U B + Urad , the radiation plus magnetic energy density in the comoving frame. Fig. 3 has been updated by adding 3 new BL Lac recently detected in the TeV range (see Aharonian et al. 2005a, 2005b). These new TeV BL Lacs, namely 1ES 1101 ? 232 (cf also Wolter et al. 2000), PKS 2005 ? 489, and H 2356 ? 309, lie in the “high energy branch” de?ned by the BL Lacs previously detected in the TeV band. Interestingly, PKS 1406?076 moved toward the region of the BL Lac region, while the previous modeling – with only an upper limit in X-rays – placed this FSRQ in the region typical of these sources. The presence of two radiogalaxies in the present sample suggests some hints about the paradigm of the uni?cation of radioloud AGN (Urry & Padovani 1995), of which the γ?ray loud AGN are the subclass analyzed in this work. This paradigm and the theories on the blazar evolution (see B¨ ttcher & Dermer 2002, Cavaliere & D’Elia 2002, Maraschi & Tavecchio 2003) ?nd o analogies between BL Lac and FRI on one side and FSRQ and FRII on the other. BL Lac and FRI are evolved AGN, with low emission from the environment around the SMBH, while FSRQ and FRII are instead young sources with a rich environment.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

11

Table 5. Parameters useful to understand γ?ray loudness. Columns: (1) Source name; (2) beaming factor δ; (3) observed ?ux in the 0.4 ? 10 keV energy band [erg cm?2 s?1 ]; (4) intrinsic luminosity in the 0.4 ? 10 keV energy band [erg s?1 ]; (5) Con?dence of the EGRET detection (high > 95%; low < 95%).
Source (1) 3C 273 NGC 6251 PKS 0521 ? 365 Cen A δ (2) 6.5 ? 7 3.2 ? 3.8 1.4 ? 3 1.2 ? 1.6 F (3) ≈ 10?10 ≈ 10?12 ≈ 10?11 ≈ 10?10 L (4) ≈ 1046 ≈ 1043 ≈ 1042 ≈ 1041 Conf. (5) high low low high

With reference to the γ?ray propagation from the source to the observer, one of the most important factor is the beaming factor δ, that allow high energy photons to escape from the source without disappearing in pair production. From the SED of the γ?ray loud AGN reported in Ghisellini et al. (1998, 2005), Chiaberge et al. (2001, 2003), Foschini et al. (2005), complemented and con?rmed by this work, we see that the δ values for almost all the EGRET detected AGN are above 10, with only a few exceptions. 3C 273 with δ = 6.5 ? 7 and Cen A with δ = 1.2 ? 1.6 are also the only AGN with δ < 10 detected by EGRET with high con?dence. PKS 0521 ? 365 (δ = 1.4 ? 3) and NGC 6251 (δ = 3.2 ? 3.8) have low con?dence identi?cations, that should be con?rmed. The two δ of the low con?dence detections are in between 3C 273 and Cen A, which are instead detected at high con?dence level. Therefore, there should be other reasons to explain the EGRET detections. Indeed, three of the four above sources have also the lowest intrinsic luminosities in the present sample (and also among the whole EGRET sample), but not the ?uxes (cf Table 5). This implies that the present de?nition of γ?ray loudness – that it is de?ned here simply as the detection at E > 100 MeV – is still strongly biased by the instrument sensitivity or by the distance of the source. Moreover, there is another key point still missing in this picture: γ?ray detection of FRII radiogalaxies, that are still completely missing to date even in the list of hypothetical associations.

7. Final remarks
A small sample of AGN γ?ray loud (i.e. detected by the EGRET instrument on board CGRO) observed by XMM-Newton has been analyzed in a homogeneous way and presented here. The sample is composed of 15 AGN divided into 7 FSRQ, 4 LBL, 2 HBL, and 2 FRI radiogalaxies. All the data were taken from the public archive of XMM-Newton: 46 pointings were analyzed, of which 30 are of three sources only (3C 273, Mkn 421, PKS 2155 ? 304). Despite these limitations, some useful inferences can be made. Indeed, with XMM-Newton it is possible to perform simultaneous X-ray and optical/UV observations, that can be particularly useful in blazars to place reliable constraints on the synctrotron and inverse Compton peaks in the SED. The main ?ndings can be summarized as follows: all the blazars obey the sequence suggested by Ghisellini et al. (1998) and Fossati et al. (1998). The only X-ray features found in the present sample are the emission lines of the iron complex in Cen A and in 3C 273. In the case of Cen A, the iron line at 6.4 keV (σ 80 eV) is known to be due to the transmission of radiation along the dust lane warped around Cen A, while in 3C 273, the broad iron line can be associated with the Seyfert-like component. The comparison with BeppoSAX data show a preference of the broken power law model over the single power law; the latter was often the best ?t in the BeppoSAX catalog, suggesting a long term change in the sources. The changes in the spectra of 3C 273 appear to be genuinely due to a variation in the state of the source, as well as in the case of AO 0235 ? 164, observed during an outburst. Four sources show intervening systems along the line of sight, but only one case is known to be gravitationally lensed. In the remaining three cases, it is not clear if the intervening galaxies can generate gravitational e?ects strong enough to enhance the γ?ray loudness. See Torres et al. (2003) for a discussion on this topic. The SED compiled in the present sample con?rm the model parameters already found in previous studies and increase the number of modelled sources. However, PKS 1406 ? 076 shows some particular features and deserves further investigation. Four sources appear to be the key to understand the transition (with respect to γ?ray loudness) from blazars to radiogalaxies, namely 3C 273, PKS 0521 ? 365, NGC 6251, and Cen A. The ?rst two are FSRQ with the largest jet viewing angle, which in turn means the lowest δ among the blazars. The latter two are the (only) radiogalaxies detected by EGRET. Further and more detailed studies on these four sources could give important contributions to the comprehension of the mechanisms acting to generate γ?rays, of the uni?cation models of AGN, and to the improvement of the resolution of the extragalactic γ?ray background.
Acknowledgements. LF thanks G.G.C. Palumbo, P. Grandi and M. Dadina for useful discussions and S.R. Rosen of the OM Team for useful hints in the OM data analysis. We thank also the referee, X. Barcons, for useful comments, that helped to improve the manuscript. This research has made use of the NASA’s Astrophysics Data System Abstract Service and of the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space

12

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

Administration. This work was partly supported by the European Community’s Human Potential Programme under contract HPRN-CT-200200321 and by the Italian Space Agency (ASI).

References
Aharonian F., Akhperjanian A.G., Aye K.M., et al., 2005a, A&A 436, L17 Aharonian F., Akhperjanian A.G., Bazer-Bachi A.R., et al., 2005b, submitted to Nature, (astro-ph/0508073) Bechtold J., Siemiginowska A., Aldcroft T.L., et al., 2001, ApJ 562, 133 Bergeron J. & Boiss? P., 1991, A&A 243, 344 e Birkinshaw M., Worrall D.M. & Hardcastle M.J., 2002, MNRAS 335, 142 B¨ ttcher M. & Dermer C.D., 2002, ApJ 564, 86 o B¨ ttcher M., Harvey J., Joshi M., et al., 2005, ApJ 631, 169 o Brinkmann W., Sembay S., Gri?ts R.G., et al., 2001, A&A 365, L162 Brinkmann W., Papadakis I.E., den Herder J.W.A. & Haberl F., 2003, A&A 402, 929 Cagnoni I., Nicastro F., Maraschi L., Treves A. & Tavecchio F., 2004, ApJ 603, 449 Cappi M., Matsuoka M., Comastri A., et al., 1997, ApJ 478, 492 Cavaliere A. & D’Elia V., 2002, ApJ 571, 226 Chiaberge M., Capetti A. & Celotti A., 2001, MNRAS 324, L33 Chiaberge M., Gilli R., Capetti A. & Macchetto F.D., 2003, ApJ 597, 166 Cohen R.D., Smith H.E., Junkkarinen V.T. & Burbidge E.M., 1987, ApJ 318, 577 Courbin F., Meylan G., Kneib J.P. & Lidman C., 2002, ApJ 575, 95 Courvoisier T.J.-L., Beckmann V., Bourban G., et al., 2003, A&A 411, L343 Croston J.H., Hardcastle M.J., Birkinshaw M. & Worrall D.M., 2003, MNRAS, 346, 1041 Danziger I.J., Fosbury R.A.E., Goss W.M. & Ekers R.D., 1979, MNRAS 188, 415 De Rosa A., Piro L., Tramacere A., et al., 2005, A&A 438, 121 Dickey J.M. & Lockman F.J., 1990, ARA&A 28, 215 Donato D., Sambruna R.M. & Gliozzi M., 2005, A&A 433, 1163 Edelson R., Gri?ths G., Markowitz A., et al., 2001, ApJ 554, 274 Evans D.A., Kraft R.P., Worrall D.M., et al., 2004, ApJ 612 786 Evans D.A., Worrall D.M., Hardcastle M.J., Kraft R.P. & Birkinshaw M., 2005, ApJ, accepted for publication (astro-ph/0512600) Fabian A.C., Celotti A., Iwasawa K., et al., 2001, MNRAS 323, 373 Fang T., Marshall H.L., Bryan G.L. & Canizares C.R., 2001, ApJ 555, 356 Foschini L., Chiaberge M., Grandi P., et al., 2005, A&A, 433, 515 Foschini L., Ghisellini G., Raiteri C.M., et al., 2006, Proceedings of “The X-ray Universe”, ed. A. Wilson, ESA SP-604, p. 615 (astro-ph/0510793) Fossati G., Maraschi L., Celotti A., Comastri A. & Ghisellini G., 1998, MNRAS 299, 433 Ghisellini G., 2004, Nuclear Physics B Proceedings Supplements 132, 76 Ghisellini G., Celotti A., Fossati G., Maraschi L. & Comastri A., 1998, MNRAS 301, 451 Ghisellini G., Celotti A. & Costamante L., 2002, A&A 386, 833 Ghisellini G., Tavecchio F. & Chiaberge M., 2005, A&A 432, 401 Giommi P., Capalbi M., Fiocchi M., et al., 2002, Proceedings of Blazar Astrophysics with BeppoSAX and Other Observatories, eds P. Giommi, E. Massaro, G.G.C. Palumbo, ASI Science Data Center and ESA-ESRIN, p. 63 (astro-ph/0209596) Giommi P., Colafrancesco S., Cavazzuti E., Perri M., Pittori C., 2006, A&A 445, 843 Gliozzi M., Sambruna R.M., Brandt W.N., et al., 2004, A&A 413, 139 Grandi P., Fiocchi M., Perola C.G., et al., 2003, ApJ 593, 160 Grandi P. & Palumbo G.G.C., 2004, Science 306, 998 Grandi P., Malaguti G. & Fiocchi M., 2005, ApJ, accepted for publication (astro-ph/0511784) Guainazzi M., Grandi P., Comastri A. & Matt G., 2003, A&A 410, 131 Hardcastle M.J., Worrall D.M. & Birkinshaw M., 1999, MNRAS 305, 246 Hartman R.C., Bertsch D.L., Bloom S.D., et al., 1999, ApJS 123, 79 Hutchison J.M., Cawthorne T.V. & Gabuzda D.C., 2001, MNRAS 321, 525 Jansen F., Lumb D., Altieri B., et al., 2001, A&A 365, L1 Junkkarinen V.T., Cohen R.D., Beaver E.A., et al., 2004, ApJ 614, 658 Kirsch M., 2005. EPIC status of calibration and data analysis. XMM-SOC-CAL-TN-0018, v. 2.4, 11 February 2005. Kuiper L., Hermsen W., Verbunt F., et al., 2000, A&A 359, 615 Lamb R.C. & Macomb D.J., 1997, ApJ 488, 872 Lane W., Smette A., Briggs F., et al., 1998, AJ 116, 26 Landt H., Perlman E.S. & Padovani P., 2006, ApJ, 637, 183 Lidman C., Courbin F., Meylan G., et al., 1999, ApJ 514, L57 Madejski G., Takahashi T., Tashiro M., et al., 1996, ApJ 459, 156 Malizia A., Bassani L., Dean A.J., et al., 2000, ApJ 531, 642 Maraschi L., Ciapi A., Fossati G., Tagliaferri G. & Treves A., 1995a, ApJ 443, 29 Maraschi L., Fossati G., Tagliaferri G. & Treves A., 1995b, ApJ 443, 578

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei

13

Maraschi L., Tavecchio F., Cagnoni I., et al., 2002, In: New Visions of the X-ray Universe in the XMM-Newton and Chandra era. ESA-ESTEC, 2001 (astro-ph/0202418) Maraschi L. & Tavecchio F., 2003, ApJ 593, 667 Mason K.O., Breeveld A., Much R., et al., 2001, A&A 365, L36 Mathur S. & Nair S., 1997, ApJ 484, 140 Molendi S. & Sembay S., 2003, Assessing the EPIC spectral calibration in the hard band with 3C 273 observation. XMM-SOC-CAL-TN-0036, March 2003. Mukherjee R., Halpern J., Mirabal N. & Gotthelf E.V., 2002, ApJ 574, 693 Oshima T., Mitsuda K., Ota N., et al., 2001, ApJ 551, 929 Padovani P., Perlman E.S., Landr H., Giommi P. & Perri M., 2003, ApJ 588, 128 Page K.L., Turner M.J.L., Done C., et al., 2004, MNRAS 349, 57 Pei Y.C., Fall S.M. & Bechtold J., 1991, ApJ 378, 6 Pettini M., Smith L.J., Hunstead R.W. & King D.L., 1994, ApJ 426, 79 Perlman E., Madejski G., Georganopoulos M., et al., 2005, ApJ 625, 727 Pian E., Falomo R., Ghisellini G., et al., 1996, ApJ 459, 169 Pian E., Vacanti G., Tagliaferri G., et al., 1998, ApJ 492, L17 Pian E., Foschini L., Beckmann V., et al., 2005, A&A 429, 427 Piconcelli E. & Guainazzi M., 2005, A&A 442, L53 Protassov R., van Dyk D.A., Connors A., et al., 2002, ApJ 571, 545 Raiteri C.M., Villata M., Aller H.D., et al., 2001, A&A 377, 396 Raiteri C.M., Villata M., Tosti G., et al., 2003, A&A 402, 151 Raiteri C.M., Villata M., Ibrahimov M.A., et al., 2005, A&A 438, 39 Raiteri C.M., Villata M., Kadler M., et al., 2006, A&A, accepted for publication (astro-ph/0603364) Ravasio M., Tagliaferri G., Ghisellini G. & Tavecchio F., 2004, A&A 424, 841 Reynolds C.S. & Nowak M.A., 2003, Physics Reports 377, 389 Sambruna R.M., Maraschi L. & Urry C.M., 1996, ApJ 463, 424 Sambruna R.M., Gliozzi M., Donato D., et al., 2004, A&A 414, 885 Sbarufatti B., Treves A., Falomo R., 2005, ApJ 635, 173 Sembay S., Edelson R., Markowitz A., et al., 2002, ApJ 574, 634 Siebert J., Brinkmann W., Drinkwater M.J., et al., 1998, MNRAS 301, 261 Siemiginowska A., Bechtold J., Aldcroft T.L., et al. 2002, ApJ 570, 543 Snowden S., Still M., Harrus I. et al., 2004. An introduction to XMM-Newton data analysis. Version 2.01, 23 July 2004. Sowards-Emmerd D., Romani R.W. & Michelson P., 2003, ApJ 590, 109 Sowards-Emmerd D., Romani R.W., Michelson P. & Ulvestad J.S., 2004, ApJ 609, 564 Sowards-Emmerd D., Romani R.W., Michelson P., et al., 2005, ApJ 626, 95 Stanghellini C., O’Dea C.P., Dallacasa D., et al., 1998, A&ASS 131, 303 Stanghellini C., 2003, PASA 20, 118 Stickel M. & K¨ hr H., 1993, A&AS 100, 395 u Strong A.W., Moskalenko I.V., Reimer O., 2004, ApJ 613, 956 Str¨ der L., Briel U., Dennerl K., et al., 2001, A&A 365, L18 u Swanenburg B.N., Hermsen W., Bennett K., et al., 1978, Nature 275, 298 Tagliaferri G., Ghisellini G., Giommi P., et al., 2000, A&A 354, 431 Tagliaferri G., Ghisellini G. & Ravasio M., 2002, Proceedings of Blazar Astrophysics with BeppoSAX and Other Observatories, eds P. Giommi, E. Massaro, G.G.C. Palumbo, ASI Science Data Center and ESA-ESRIN, p. 11 (astro-ph/0207017) Tagliaferri G., Ravasio M., Ghisellini G., et al., 2003, A&A 400, 477 Tavecchio F., Maraschi L., Ghisellini G., et al., 2000, ApJ 543, 535 Tavecchio F., Maraschi L., Ghisellini G., et al., 2002, ApJ 575, 137 Ter¨ sranta H., Wiren S., Koivisto P., Saarinen V., & Hovatta T., 2005, A&A 440, 409 a Thompson D.J., Bertsch D.L., Dingus B.L., et al., 1995, ApJS 101, 259 Tingay S.J. & Edwards P.G., 2002, AJ 124, 652 Torres D.F., Romero G.E., Eiroa E.F., Wambsganss J. & Pessah M.E., 2003, MNRAS 339, 335 Turner M.J., Abbey A., Arnaud M., et al., 2001, A&A 365, L27 Ulrich M.-H., Maraschi L., & Urry C.M., 1997, ARA&A 35, 445 Urry C.M. & Padovani P., 1995, PASP 107, 803 von Montigny C., Bertsch D.L., Chiang J., et al., 1995, ApJ 440, 525 Wagner S.J., & Witzel A., 1995, ARA&A 33, 163 Wiklind T. & Combes F., 1996, Nature 379, 11 Wolfe A.M., Gawiser E. & Prochaska J.X., 2005, ARA&A 43, 861 Wolter A., Tavecchio F., Caccianiga A., Ghisellini G. & Tagliaferri G., 2000, A&A 357, 429 Zhang Y.H., Treves A., Celotti A., Qin Y.P. & Bai J.M., 2005, ApJ 629, 686

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 1

Online Material

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 2

Appendix A: Notes on individual sources, tables and SED
We report in this Appendix the tables with the Observation log (Table A.1), the ?t with the simple power law model (Table A.2), the broken power law model (Table A.3), the additional ?ts for 3C 273 (Table A.4 and A.5) and Cen A (Table A.6) and the magnitudes with di?erent ?lters of the Optical Monitor (Table A.7). Some notes on the individual sources and the SED of the sources not reported in Sect. 6 (Fig. A.1 and A.2) complete this Appendix. 3EG J0222+4253 (0219+428, 3C 66A): The counterpart of this EGRET source is the BL Lac 0219 + 428, although Kuiper et al. (2000) have proposed that the ?ux below 300 MeV is signi?cantly contaminated by the nearby pulsar PSR J0218+4232. 3C 66A was extensively monitored from radio to very high γ?rays during 2003?2004 (B¨ ttcher et al. 2005) and no signi?o cant X-ray variability was detected. The present XMM-Newton data set was a?ected by a high background, but it is possible to use more than 70% of the observation. The ?t of the spectrum of this BL Lac object with an absorbed power law model is acceptable (see Table A.2), although there are residuals for energies greater than 4 keV. The broken power law model provides the best ?t with a con?dence > 99.99% with a f-test. This observation has been studied by Croston et al. (2003) and the ?t with a simple power law absorbed by the Galactic NH is consistent with the present results. However, signi?cant di?erences are present in the ?t with the broken power law model: this can be understood by taking into account that in the present analysis a more conservative selection of events has been used (for example, this can be clearly inferred by comparing the PN exposures: 11 ks in the present analysis vs 15 ks in the analysis of Croston et al., 2003). 3EG J0237+1635 (AO 0235+164): The data set analyzed here refers to an observation performed in 2002 and shows no signs of high background. The epatplot task of XMM SAS shows a slight excess of double pixels events and a corresponding de?cit of single pixel events (pile-up), that can be easily suppressed by removing the inner region with 5′′ radius. The ?t can improve signi?cantly by reducing the energy band to 1 ? 10 keV, that is by removing the energy band that can be affected by the absorption and the ?t reported in Tables A.2 and A.3 refer to this case. The best ?t is still obtained with the broken power law model with the absorption column ?xed to the Galactic value, although the Γ1 is slightly harder than the above mentioned case. 3EG J0530-3626 (PKS 0521-365): The blazar PKS 0521 ? 365 has been associated with the EGRET source in the Second Catalog (Thompson et al. 1995), but a stronger detection in the Cycle 4 placed this blazar outside the 99.99% probability contours. Sowards-Emmerd et al. (2004) suggested that the counterpart of 3EG J0530 ? 3626 source could be another radio source (PMN J0529?3555) of unknown nature. This can be another case of a possible double source not resolved by EGRET (like, e.g. 3EG J0222 + 4253), an interesting target worth ob-

serving with GLAST. In the present work, we keep as valid the association with PKS 0521 ? 365. The XMM-Newton observation is analyzed here for the ?rst time. There is no evidence of high background, but the epatplot task shows that the data are a?ected by pile-up. This blazar has a small jet (6′′ ), visible in radio (see Tingay & Edwards 2002 for a description of the parsec scale structure), and optical wavelenghts (Danziger et al. 1979). Chandra detected a jet-like feature of 2′′ -size, spatially coincident with the optical and radio structure (Birkinshaw et al. 2002). Hardcastle et al. (1999) and Birkinshaw et al. (2002) reported also the presence of extended emission that can be ?tted with thermal plasma model with kT = 1.6 keV and Z⊙ = 0.05. Given the size of the PSF of EPIC camera on board XMM-Newton (12′′ ? 15′′ HEW, see Jansen et al. 2001), the above mentioned structures are not resolved. Attempts to ?t the low energy part with a thermal plasma model (mekal or raymond models in xspec, not reported in the Tables), resulted in a χ2 ≈ 1.08 improved with respect to the simple power law ? (?χ2 = 84 for a decrease of 2 dof), but still worse than the broken power law model. The broken power law model absorbed by the Galactic column provides the best ?t to the present data, consistent with the results obtained by Einstein, EXOSAT (Pian et al. 1996), ROSAT (Pian et al. 1996, Hardcastle et al. 1999), BeppoSAX (Tavecchio et al. 2002). 3EG J0721+7120 (S5 0716+714): This BL Lac object is known for its extreme variability, also at intraday time scales (cf Wagner & Witzel 1995), and it is extensively monitored by optical and radio ground telescopes (see, e.g. Raiteri et al. 2003). The present observation was performed simultaneously with the end of a ToO of INTEGRAL, triggered after an optical giant ?are (Pian et al. 2005). The optical lightcurve started to increase at the end of March 2004, and the trigger was activated on 27 March. The INTEGRAL ToO was performed from 2 to 7 April, when the source activity was already declining. The XMM-Newton observation covered instead the period 4 ? 5 April. The X-ray lightcurve shows a clear decrease of the ?ux as the observation proceeded, con?rming that the source was observed during the tail of the ?are. More details on this XMM-Newton data set are presented in a separate paper (Foschini et al., submitted). 3EG J0845+7049 (S5 0836+710): This ?at-spectrum radio quasar has one of the highest redshifts among the objects in the present sample. The XMM-Newton data, published here for the ?rst time, show high background. Past analyses of ROSAT and ASCA data by Cappi et al. (1997) and BeppoSAX data by Tavecchio et al. (2000) have shown a spectrum with a hard photon index (Γ ≈ 1.3), with acceptable ?t also with a broken power law, but some inconsistencies in the value of the absorption. The present data are best ?tted with a simple power law model absorbed by the Galactic column plus an additional absorber at the redshift of the quasar with a value (1.4±0.3)×1021 cm?2 , consistent with the ASCA and BeppoSAX values. This is also consistent with Chandra observations by Fang et al. (2001), who found Γ = 1.388 ± 0.012 (0.5 ? 8 keV), but a slightly lower absorption NH = (7.0 ± 1.2) × 1020 cm?2 (not redshifted). The photon index is also consistent with the

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 3

Table A.1. XMM-Newton Observation Log. Columns: (1) Source name; (2) Observation Identi?er; (3) Date of the observation [DD-MM-YYYY]; (4,5,6) Observing mode of MOS1, MOS2, and PN, respectively [FF: Full Frame; SW: Small Window; TIMING: timing mode, without imaging] with the e?ective exposure time [ks]; (7) Angular distance from the boresight [arcsec].
Name (1) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714 S5 0836 + 710 Mkn 421a ObsID (2) 0002970201 0110990101 0065760201 0150495601 0112620101 0099280201 0099280301 0099280401 0136540101 0158970101 0162960101 0112850201 0104860501 0126700301 0126700601 0126700701 0126700801 0136550101 0112770101 0112770201 0112770601 0112770801 0136550501 0112770701 0112771001 0112770501 0112771101 0136550801 0093650201 0093650301 0147670201 0151590101 0151590201 0056340201 0204580201 0204580301 0204580401 0124930201 0080940101 0080940301 0124930301 0124930501 0124930601 0158960101 0158960901 0158961001 Date (3) 05 ? 02 ? 2002 10 ? 02 ? 2002 09 ? 10 ? 2002 04 ? 04 ? 2004 12 ? 04 ? 2001 01 ? 11 ? 2000 13 ? 11 ? 2000 14 ? 11 ? 2000 08 ? 05 ? 2001 01 ? 06 ? 2003 10 ? 12 ? 2003 01 ? 07 ? 2002 26 ? 06 ? 2002 13 ? 06 ? 2000 15 ? 06 ? 2000 15 ? 06 ? 2000 17 ? 06 ? 2000 13 ? 06 ? 2001 16 ? 12 ? 2001 22 ? 12 ? 2001 07 ? 07 ? 2002 17 ? 12 ? 2002 05 ? 01 ? 2003 05 ? 01 ? 2003 18 ? 06 ? 2003 08 ? 07 ? 2003 14 ? 12 ? 2003 30 ? 06 ? 2004 02 ? 02 ? 2001 06 ? 02 ? 2002 31 ? 01 ? 2003 05 ? 07 ? 2003 10 ? 08 ? 2003 26 ? 03 ? 2002 10 ? 03 ? 2004 24 ? 03 ? 2004 05 ? 04 ? 2004 31 ? 05 ? 2000 19 ? 11 ? 2000 20 ? 11 ? 2000 30 ? 11 ? 2001 24 ? 05 ? 2002 29 ? 11 ? 2002 23 ? 11 ? 2003 22 ? 11 ? 2004 23 ? 11 ? 2004 MOS1 (4) FF(10) FF(19) SW(31.3) SW(31) SW(23.8) SW SW SW SW TIMING SW FF(13.9) FF(33.1) SW SW SW SW SW TIMING TIMING TIMING TIMING SW TIMING TIMING TIMING TIMING SW FF(22.8) FF(13.2) FF(13.5) FF(7.8) FF(5.6) FF(18.0) FF(7.0) FF(31.0) FF(18.5) TIMING TIMING TIMING SW SW SW SW SW SW MOS2 (5) FF(10) FF(19) FF(31.3) FF(31) SW(23.8) SW SW SW SW SW SW FF(13.7) FF(32.6) SW SW SW SW SW SW SW SW SW SW SW SW SW SW SW FF(22.8) FFb FF(13.5) FF(10.0) FF(5.8) FF(18.0) FF(7.6) FF(31.0) FF(18.5) SW SW SW SW SW SW SW SW SW PN (6) FF(11.0) FF(15.0) FF(27.1) TIMING FF(24.6) SW(24.2) SW(25.6) SW(23.4) SW(25.7) SW(25.3) SW(17.4) FF(10.7) FF(26.7) SW(39.7) SW(20.8) SW(21.0) SW(42.5) SW(62.0) SW(3.5) SW(3.5) SW(3.5) SW(3.5) SW(6.0) SW(3.5) SW(3.9) SW(5.6) SW(5.9) SW(13.9) FF(16.8) FF(7.9) FF(10.9) FF(7.3) FF(3.6) FF(8.0) FF(2.8) FF(27.0) FF(13.0) SW(41.6) SW(40.2) SW(40.8) SW(31.2) SW(22.3) SW(39.8) SW(18.7) SW(20.0) SW(28.0) Position (7) 366 77 65 69 14 6.5 7.1 119.7 4.7 6.1 6.4 68 472 10 9 8 9 5 54.8 54.5 78.7 55.0 5 58.8 79.1 78.0 57.2 3 9 16 68 68 69 69 67 68 68 9 10 16 15 6 2 5 6 6

PKS 1127 ? 145 ON 231 3C 273a

Cen A PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211

PKS 2155 ? 304a

a b

Only PN data have been analyzed, because the high ?ux of the source caused strong pile-up. MOS2 not used, because of a series of bad pixels in the source PSF.

value of Γ = 1.3 ± 0.3 in the 20 ? 100 keV energy band measured by INTEGRAL (Pian et al. 2005) and the value of Γ = 1.1 ± 0.3 measured by CGRO with OSSE and BATSE instruments (Malizia et al. 2000). The broken power law gives also a good ?t, although with no improvement with respect to

the simple power law model, but does not require the additional absorber. 3EG J1104+3809 (Mkn 421): This is one of three sources in the present sample that has been extensively observed, being

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 4

Table A.2. Fit results with single powerl law model. Columns: (1) Source name; (2) absorbing column density [1020 cm?2 ]; (3) photon index of the power law; (4) Normalization of the power law model [10?2 ph cm?2 s?1 keV?1 at 1 keV]; (5) Reduced χ2 and degrees of freedom; (6) observed ?ux in the 0.4 ? 10 keV energy band [10?11 erg cm?2 s?1 ]; (7) intrinsic luminosity in the 0.4 ? 10 keV energy band rest frame [1045 erg s?1 ]. The uncertainties in the parameter estimates are at the 90% con?dence limits for 1 parameter. For NH = Gal, it means that the absorption column has been ?xed to the Galactic value.
Name (1) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714a S5 0836 + 710b Mkn 421c NH (2) Gal. Gal. Gal. Gal. 14 ± 3 2.3 ± 0.3 3.5 ± 0.2 3.1 ± 0.2 3.1 ± 0.2 4.6 ± 0.2 3.7 ± 0.2 12+2 ?1 2.5 ± 0.6 Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. 1400 ± 200 1800 ± 300 6.7 ± 0.9 Gal. Gal. +0.8 11.6?0.7 350 ± 50 240 ± 10 220 ± 10 1.6 ± 0.2 2.2 ± 0.2 1.9 ± 0.3 3.2 ± 0.2 1.4 ± 0.3 3.0 ± 0.3 Gal. 2.5 ± 0.5 2.9 ± 0.4 Γ (3) 2.60 ± 0.04 2.28 ± 0.02 1.847 ± 0.009 2.58 ± 0.02 1.379 ± 0.007 2.61 ± 0.01 2.438 ± 0.006 2.457 ± 0.006 2.373 ± 0.007 2.73 ± 0.01 2.43 ± 0.01 1.31 ± 0.02 2.77 ± 0.04 1.829 ± 0.004 1.819 ± 0.005 1.812 ± 0.005 1.8 ± 0.2 1.9 ± 0.2 1.87 ± 0.01 1.82 ± 0.01 1.89 ± 0.01 1.98 ± 0.01 2.019 ± 0.009 2.02 ± 0.01 1.950 ± 0.009 1.963 ± 0.008 1.901 ± 0.009 1.899 ± 0.006 2.2 ± 0.1 2.3 ± 0.2 1.80 ± 0.04 1.58 ± 0.07 1.6 ± 0.1 1.94 ± 0.03 +0.03 1.17?0.05 1.14 ± 0.02 1.13 ± 0.03 2.592 ± 0.009 2.81 ± 0.01 2.86 ± 0.01 2.868 ± 0.009 2.70 ± 0.01 2.89 ± 0.01 2.96 ± 0.01 3.08 ± 0.02 2.98 ± 0.02 A (4) 0.183 ± 0.007 1.22 ± 0.04 0.231 ± 0.002 0.294 ± 0.003 2.62 ± 0.04 7.74 ± 0.09 26.3 ± 0.1 25.6 ± 0.1 18.3 ± 0.1 13.5 ± 0.1 16.3 ± 0.1 0.34 ± 0.01 0.186 ± 0.005 2.788 ± 0.007 2.68 ± 0.01 2.603 ± 0.009 2.599 ± 0.004 3.634 ± 0.004 4.18 ± 0.03 3.97 ± 0.03 3.09 ± 0.02 4.67 ± 0.03 3.92 ± 0.02 3.99 ± 0.03 4.78 ± 0.03 4.13 ± 0.02 3.07 ± 0.02 2.64 ± 0.01 22+7 ?5 29+14 ?9 0.171 ± 0.009 +0.007 0.060?0.006 0.0560.009 ?0.007 0.134 ± 0.004 +0.07 0.62?0.08 0.54 ± 0.03 0.47 ± 0.03 5.12 ± 0.04 4.49 ± 0.04 3.61 ± 0.04 8.23 ± 0.06 3.68 ± 0.05 2.81 ± 0.04 2.40 ± 0.01 2.80 ± 0.05 3.77 ± 0.05 χ2 /dof ? (5) 1.29/293 0.99/618 1.15/1202 1.74/419 1.07/2272 0.97/1026 1.20/1491 1.29/1430 1.12/1447 1.16/1178 1.03/1307 1.01/890 1.07/653 1.82/1598 1.71/1401 1.53/1394 2.08/1647 4.22/1711 1.24/802 1.15/806 1.08/680 1.21/790 1.37/899 1.23/711 1.27/826 1.12/908 1.33/877 1.32/1160 0.97/668 0.97/340 0.93/609 0.79/90 0.94/49 1.04/812 0.90/600 1.04/1662 1.04/1168 1.02/1266 1.00/1067 1.03/991 1.01/1131 0.99/886 0.87/880 0.95/614 0.88/625 1.10/831 F (6) 0.21 0.96 1.2 1.0 4.8 26.0 90.6 88.8 66.1 40.6 56.0 1.1 0.50 12.2 11.8 11.5 11.5 14.5 17.4 17.5 12.7 17.7 14.3 14.5 18.5 15.8 12.5 10.7 16.9 18.5 0.43 0.10 0.094 0.577 1.42 1.34 1.22 14.4 11.6 9.32 20.2 10.0 6.90 6.17 6.83 9.14 L (7) 2.6 66 9.0 × 10?4 > 14.2 902 0.59 2.12 2.05 1.51 1.02 1.31 59.2 0.16 8.4 8.2 8.0 8.0 10.2 12.2 12.1 8.90 12.6 10.3 10.4 13.2 11.2 8.77 7.54 3.0 × 10?4 3.3 × 10?4 5.1 10.5 9.78 9.6 × 10?3 322 290 261 5.71 4.87 3.90 8.89 4.02 3.03 2.60 3.04 4.07

PKS 1127 ? 145d ON 231 3C 273

Cen Ae PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211d

PKS 2155 ? 304c

a b c d e

Only MOS1+MOS2, since the PN was set in TIMING. Lower limit of luminosity calculated for z = 0.5. Additional absorber placed at the redshift of the source (wa*zwa(zpo) model in xspec). Fit in the 0.6 ? 10 keV energy range and extrapolation to 0.4 keV for ?ux and luminosity calculations. Additional redshifted absorber (wa*zwa(zpo) model in xspec) along the line of sight. Fit in the 4 ? 10 keV energy range, without 6 ? 8 keV energy band, because of complex features in the low energy part and in the iron emission line complex.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 5

Table A.3. Fit results with the broken power law model. Columns: (1) Source name; (2) absorbing column density [1020 cm?2 ]; (3) low energy photon index of the power law; (4) high energy photon index; (5) Break energy [keV]; (6) Normalization of the power law model [10?2 ph cm?2 s?1 keV?1 at 1 keV]; (7) Reduced χ2 and degrees of freedom; (8) observed ?ux in the 0.4 ? 10 keV energy band [10?11 erg cm?2 s?1 ]; (9) intrinsic luminosity in the 0.4 ? 10 keV energy band rest frame [1045 erg s?1 ]; (10) Ftest probability [%]; ?t with probability below 90% are not reported. The uncertainties in the parameter estimates are at the 90% con?dence limits for 1 parameter. For NH = Gal, it means that the absorption column has been ?xed to the Galactic value.
Name (1) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714a S5 0836 + 710 Mkn 421b NH (2) Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. 12+2 ?1 Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Gal. Γ1 (3) +0.12 2.91?0.08 2.33 ± 0.04 1.95 ± 0.03 2.70 ± 0.02 1.25 ± 0.03 2.573 ± 0.008 2.340 ± 0.004 +0.004 2.378?0.005 2.312 ± 0.007 +0.01 2.59?0.02 2.300 ± 0.007 +0.08 1.40?0.05 2.74 ± 0.02 1.943 ± 0.009 1.97 ± 0.01 1.96 ± 0.01 1.92 ± 0.01 2.121 ± 0.007 2.00 ± 0.02 1.94 ± 0.03 1.97 ± 0.03 2.09 ± 0.02 2.16 ± 0.02 2.13 ± 0.02 2.05 ± 0.02 2.02 ± 0.02 2.02 ± 0.02 +0.02 2.00?0.01
+0.08 1.57?0.24

PKS 1127 ? 145c ON 231 3C 273

Γ2 (4) +0.10 2.23?0.09 2.1 ± 0.1 1.74 ± 0.03 +0.08 1.98?0.09 1.375 ± 0.007 +0.17 2.79?0.08 2.73 ± 0.04 +0.08 2.96?0.10 +0.07 2.68?0.05 2.81 ± 0.02 2.51 ± 0.02 1.22 ± 0.06 +0.3 2.3?0.4 1.697 ± 0.009 1.65 ± 0.01 1.67 ± 0.01 1.654 ± 0.009 1.676 ± 0.008 1.72 ± 0.02 1.67 ± 0.03 +0.03 1.79?0.04 1.81 ± 0.03 +0.02 1.83?0.03 +0.04 1.81?0.05 1.76 ± 0.03 +0.03 1.82?0.07 1.70 ± 0.03 1.76 ± 0.02

Ebreak (5) 1.3 ± 0.2 +0.7 3.3?0.5 +0.3 1.5?0.2 +0.2 2.3?0.1 +0.15 1.01?0.08 +1.0 4.2?0.7 4.0 ± 0.2 +0.2 4.7?0.3 4.4 ± 0.3 +0.1 2.2?0.2 2.2 ± 0.1 +1.0 2.7?0.8 +0.9 4.2?1.1 1.48 ± 0.06 1.43 ± 0.08 1.35 ± 0.08 1.37 ± 0.05 1.45 ± 0.03 +0.2 1.4?0.1 1.5 ± 0.2 +0.4 1.4?0.2 1.6 ± 0.1 +0.2 1.4?0.1 +0.3 1.7?0.2 1.8 ± 0.2 +0.6 1.9?0.3 +0.2 1.7?0.1 +0.2 1.5?0.1

A (6) 0.067 ± 0.003 0.276 ± 0.006 0.227 ± 0.003 0.293 ± 0.003 0.529 ± 0.006 6.97 ± 0.02 22.9 ± 0.4 22.6 ± 0.4 16.4 ± 0.7 +0.7 11.5?0.8 14.1 ± 0.4 +0.006 0.127?0.004 0.140 ± 0.002 2.073 ± 0.008 1.97 ± 0.01 1.91 ± 0.01 1.911 ± 0.008 2.604 ± 0.009 3.06 ± 0.03 2.95 ± 0.03 2.29 ± 0.03 3.39 ± 0.03 2.80 ± 0.03 2.88 ± 0.03 3.51 ± 0.03 3.05 ± 0.02 2.26 ± 0.02 1.94 ± 0.01

χ2 /dof ? (7) 1.05/291 0.96/616 1.05/1200 1.06/417 1.07/2271 0.96/1025 1.06/1490 1.02/1429 0.99/1446 1.08/1177 0.99/1306 1.00/888 1.07/652 1.13/1596 1.03/1399 1.00/1392 0.97/1645 1.10/1709 1.00/800 0.95/804 1.01/678 0.97/788 0.99/897 0.96/709 1.02/824 1.00/906 1.02/875 1.03/1158

F (8) 0.24 0.99 1.3 1.1 4.8 26.1 91.3 89.1 66.7 41.8 56.3 1.2 0.51 12.7 12.5 12.1 12.2 15.7 18.4 18.4 13.1 18.6 15.1 15.4 19.5 16.4 13.2 11.2

L (9) 3.0 69 9.4 × 10?4 > 15.7 860 0.578 2.00 1.95 1.45 0.929 1.23 61.0 0.16 8.7 8.6 8.3 8.4 11.0 12.7 12.6 9.09 13.2 10.8 10.9 13.7 11.6 9.20 7.8

Cen A PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211 Gal. 1.78 ± 0.03 1.0 ± 0.3
+0.012 0.077?0.003

0.92/608

0.43

4.8

PKS 2155 ? 304b

14 ± 1 62+5 ?6 62 ± 3 64 ± 3 Gal. Gal. Gal. Gal. Gal.

+0.08 2.11?0.06 +0.08 0.93?0.10 +0.06 0.89?0.09 1.06 ± 0.04 2.588 ± 0.005 +0.007 2.773?0.006 2.845 ± 0.006 2.783 ± 0.006 2.75 ± 0.03

1.78 ± 0.07 +0.10 1.42?0.05 1.29 ± 0.04 +0.12 1.41?0.09 +0.18 2.73?0.08 +0.06 2.88?0.03 3.0 ± 0.1 2.96 ± 0.04 +0.01 2.70?0.02

+0.3 2.5?0.4 +0.4 3.6?0.5 3.1 ± 0.3 +0.5 4.7?0.4 +1.4 4.4?1.1 +0.7 2.8?0.4 +0.8 4.3?1.2 +0.3 2.4?0.2 +0.5 1.0?0.1

0.141 ± 0.004 0.11 ± 0.01 +0.06 0.103?0.08 0.108 ± 0.06 3.861 ± 0.008 3.246 ± 0.007 2.620 ± 0.007 5.75 ± 0.01 2.74 ± 0.02

1.00/810 0.83/598 0.97/1660 1.00/1166 1.01/1265 0.98/1066 1.02/990 0.97/1130 0.99/885

0.59 1.38 1.31 1.20 14.3 11.6 9.31 20.3 10.0

0.010 270 243 248 5.70 4.74 3.86 8.35 4.10

Ftest (10) > 99.99 > 99.99 99.97 > 99.99 99.97 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 99.58 94.50 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 92.31 81.60 56.30 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 > 99.99 99.67 > 99.99 96.01 89.10 85.00

a b c

Only MOS1+MOS2, since the PN was set in TIMING. Lower limit of luminosity calculated for z = 0.5. Fit in the 0.6 ? 10 keV energy range and extrapolation to 0.4 keV for ?ux and luminosity calculations. Additional redshifted absorber (wa*zwa(bknpo) model in xspec) along the line of sight.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 6

Table A.4. Additional ?t results for 3C 273 (3EG J1229 + 0210) with the model composed of a black body plus a power law (wa(zbb+zpo)), absorbed with the Galactic column density. Columns: (1) Temperature [keV]; (2) Normalization of the black body model [10?4 L39 /D2 , where L39 is the source luminosity in units of 1039 erg/s and D10 is the source distance in units of 10 10 kpc]; ; (3) Photon index; (4) Normalization of the power law model [10?2 ph cm?2 s?1 keV?1 at 1 keV]; (5) Reduced χ2 and degrees of freedom; (6) observed ?ux in the 0.4 ?10 keV energy band [10?11 erg cm?2 s?1 ]; (7) intrinsic luminosity in the 0.4 ?10 keV energy band rest frame [1045 erg s?1 ]. The uncertainties in the parameter estimates are at the 90% con?dence limits for 1 parameter.
kT (1) 0.143 ± 0.004 0.143 ± 0.004 +0.006 0.138?0.004 0.137 ± 0.004 0.140 ± 0.002 +0.007 0.152?0.008 0.153 ± 0.009 0.16 ± 0.01 +0.007 0.149?0.008 0.136 ± 0.006 0.148 ± 0.007 +0.007 0.146?0.008 0.157 ± 0.008 0.146 ± 0.007 0.151 ± 0.005 Azbb (2) 1.48 ± 0.07 +0.09 1.82?0.10 +0.11 1.58?0.08 +0.06 1.76?0.10 3.47 ± 0.07 2.8 ± 0.3 2.3 ± 0.3 1.4 ± 0.3 +0.3 3.1?0.4 2.9 ± 0.2 3.1 ± 0.3 3.0 ± 0.3 2.0 ± 0.3 2.1 ± 0.2 1.6 ± 0.1 Γ (3) 1.714 ± 0.005 1.668 ± 0.007 +0.006 1.685?0.008 1.669 ± 0.009 1.701 ± 0.005 1.72 ± 0.02 1.68 ± 0.02 +0.01 1.79?0.02 1.82 ± 0.02 1.86 ± 0.01 1.84 ± 0.02 +0.01 1.80?0.02 +0.009 1.852?0.017 1.74 ± 0.02 1.762 ± 0.008 Apl (4) 2.47 ± 0.02 2.29 ± 0.03 +0.02 2.28?0.04 2.25 ± 0.02 2.93 ± 0.02 3.54 ± 0.09 3.42 ± 0.09 2.76 ± 0.08 4.0 ± 0.1 3.36 ± 0.07 3.33 ± 0.09 4.13 ± 0.09 3.67 ± 0.08 2.61 ± 0.06 2.28 ± 0.04 χ2 /dof ? (5) 1.17/1596 1.06/1399 1.01/1392 1.02/1645 1.23/1709 0.99/800 0.95/804 1.00/678 0.98/788 1.02/897 0.95/709 1.04/824 0.99/906 1.04/875 1.00/1158 F (6) 12.6 12.5 12.1 12.2 15.7 18.4 18.3 13.1 18.6 15.0 15.3 19.4 16.3 13.1 11.2 L (7) 8.7 8.6 8.3 8.3 10.9 12.7 12.5 9.06 13.1 10.7 10.9 13.6 11.5 9.13 7.8

Table A.5. Detections and upper limits on the equivalent width of iron lines for 3C 273 (3EG J1229 + 0210). The continuum is best ?tted with the broken power law model with Galactic absorbtion reported in Table A.3. Columns: (1) Equivalent width [eV] for E = 6.4 keV and σ = 0.15 keV; (2) ?χ2 with respect to the best ?t continuum (2 dof); (3) Equivalent width [eV] for E = 6.4 keV and σ = 0.5 keV; (4) ?χ2 with respect to the best ?t continuum (2 dof); (5) Equivalent width [eV] for E = 6.7 keV and σ = 0.15 keV; (6) ?χ2 with respect to the best ?t continuum (2 dof); (7) Equivalent width [eV] for E = 6.7 keV and σ = 0.5 keV; (8) ?χ2 with respect to the best ?t continuum (2 dof); The uncertainties in the parameters and upper limits estimate are at the 90% con?dence limits for 1 parameter.
σ = 0.15 keV (1) 29+13 ?12 < 16 < 19 20+13 ?12 < 20 < 37 < 56 < 38 44+40 ?37 < 62 < 52 < 42 < 60 < 40 < 40 E = 6.4 keV ?χ2 σ = 0.5 keV (2) (3) 14.3 54.5 ± 0.1 ? < 50 ? < 62 7.0 46+33 ?15 ? 38+27 ?13 ? < 59 ? < 87 ? < 105 3.3 < 117 ? 80+49 ?67 ? < 144 ? < 111 ? < 106 ? < 97 ? 33+68 ?26 ?χ2 (4) 20.2 ? ? 16.0 16.2 ? ? ? ? 4.1 ? ? ? ? 3.8 σ = 0.15 keV (5) 18.4 ± 0.1 < 28 31 ± 19 21+13 ?14 22 ± 11 < 46 < 30 < 41 < 42 64+34 ?37 < 75 < 60 43+34 ?35 < 55 < 47 E = 6.7 keV ?χ2 σ = 0.5 keV (6) (7) 5.2 39.3 ± 0.1 ? < 53 7.2 < 64 6.4 62+15 ?36 10.8 48+13 ?28 ? < 83 ? < 95 ? < 113 ? < 133 8.6 110+44 ?83 ? < 156 ? < 112 4.1 < 92 ? < 125 ? 33+74 ?27 ?χ2 (8) 7.7 ? ? 12.5 12.0 ? ? ? ? 5.8 ? ? ? ? 3.7

a calibration target (the others are 3C 273 and PKS 2155?304). 32 observations are present in the XMM-Newton data archive, but for sake of homogeneity in the present analysis, we selected only the 6 observations with EPIC PN in small window mode. We refer the reader to the several papers published with more detailed analysis of Mkn 421 data with di?erent observ-

ing modes (e.g. Brinkmann et al. 2001, 2003, Sembay et al. 2002, Ravasio et al. 2004). 3EG J1134-1530 (PKS 1127-145): This source is in the list of Gigahertz-Peaked Sources (GPS) by Stanghellini et al. (1998), Stanghellini (2003). The present XMM-Newton data set has never been published. It shows high background, but, once

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 7

cleaned of soft-proton ?ares, it is possible to extract useful information. This source also has a 30′′ -sized jet, observed in X-rays with Chandra (Siemiginowska et al. 2002), but that is still too small for the PSF size of EPIC. 3EG J1222+2841 (ON 231): This is another EGRET source that could be composed of two or more contributions: indeed, although ON 231 is outside the global 99% probability contours of 3EG J1222 + 2841, there is a strong association with the emission at E > 1 GeV (Lamb & Macomb 1997). The present XMM-Newton data set has never been analysed and shows evidence of high background towards the end of the observation. The broken power law model provides the best ?t to the data. An absorption in addition to the Galactic column is marginally detected. The values are consistent with, although slightly di?erent to the ?t to the BeppoSAX data of an observation in 1998, when ON 231 was in outburst (Tagliaferri et al. 2000). 3EG J1229+0210 (3C 273): For historical reasons, 3C 273 is one of the most observed sources in the sky, and XMMNewton spent a lot of time observing this quasar (3C 273 is also a calibration source). We refer the reader to the works by Molendi & Sembay (2003), Courvoisier et al. (2003) and Page et al. (2004) for more details in the analysis of the XMMNewton data sets. There is a general agreement between the analysis presented here and the above cited works: just to mention one case (ObsID 0136550501), for the ?t with a simple power law model with the Galactic absorption column in the 3 ? 10 keV energy band, Courvoisier et al. (2003) reported Γ = 1.74 ± 0.03, Page et al. (2004) found Γ = 1.78 ± 0.02, and the value in the present work is Γ = 1.79 ± 0.06. However, small di?erences in the procedures should be noted. Indeed, a small excess of double pixel events and a corresponding de?cit of single pixel events can be observed with epatplot task at high energies, speci?cally above 8 keV, as already indicated by Molendi & Sembay (2003). The problem is resolved by removing the inner region of 8′′ radius: the ?t improves with negligible changes in the spectral parameters. For example, let us consider the data of the ObsID 0112770101 in the 3 ? 10 keV energy band ?t with a simple power law model with Galactic absorption. By using a circular region with 40′′ radius, the best ?t gives these values: Γ = 1.69 ± 0.05 and normalization 0.033 ± 0.003 ph cm?2 s?1 keV?1 for χ2 = 1.08 and 588 dof. ? This is to be compared with Γ = 1.73 ± 0.07 and normalization 0.036 ± 0.004 ph cm?2 s?1 keV?1 for χ2 = 0.97 and 315 ? dof in the case of annular region of extraction with inner radius 8′′ and external radius 40′′ . Therefore, since the ?t improves signi?cantly with a small increase of the error bars, the annular region of extraction has been used in the data set analyzed here. An alternative model is a power law plus a blackbody (Table A.4), where the thermal component could have a physical origin and be the hard tail of a Comptonized accretion disk. However, the values found in the present work are well above the value of 54+6 eV found with BeppoSAX, but with a more ?4 complex model (Grandi & Palumbo 2004). On the other hand, 3C 273 appeared to be in a di?erent state when observed with BeppoSAX and with XMM-Newton (see also Page et al. 2004).

3EG J1324-4314 (Cen A): Centaurus A is the nearest AGN in the sky (d = 3.84 Mpc) and has been observed twice by XMM-Newton with a delay of one year between the two observations. An analysis of these data has been published by Evans et al. (2004) and we refer the reader to that paper for more details, particularly for the extranuclear environment. But, since the purpose of the present work is to study the continuum, we performed the ?t in the 4 ? 10 keV energy band, to avoid the complex features in the low energy part of the spectrum. Moreover, to guarantee a good approximation of the continuum, we ignored the energy band 6 ? 8 keV, which is a?ected by prominent emission lines. After having ?xed the power law, the energy band of the iron complex is restored and one or two Gaussian emission lines are added to the model to complete the ?t (Table A.6). Both observations require a Gaussian emission line from FeKα, but a wing toward the high energy is present. The addition of another large line with centroid at 6.8 keV determines an improvement in the ?t of the ObsID 0093650201 (although the centroid is not well constrained), but not in the ObsID 0093650301 (it should be noted that this ObsID has less statistical power, because the MOS2 data are not useful). The iron complex found here is partially in agreement with the BeppoSAX results obtained by Grandi et al. (2003): the discrepancies refer to the FeKβ line at 7.1 keV, that is not required by the present data sets. The variability of the neutral iron line found by Grandi et al. (2003) is con?rmed also by the present data: the line ?ux changed signi?cantly between the two observations (spaced by 1 year), but increasing with the source ?ux increases, the opposite of what has been found in BeppoSAX data. However, two points temporally spaced by one year are not su?cient to claim di?erent behaviour. 3EG J1339-1419 (PKS 1334-127): This ?at-spectrum radio quasar has been poorly observed in hard X-rays: the only available observations are with ROSAT and Einstein (see, e.g., Maraschi et al. 1995a). The present analysis is also the ?rst look at the hard X-ray emission (E > 4 keV) of this source. A simple power law model, with absorption in excess of the Galactic column, provides the best ?t. The photon index is in the middle of the values from Einstein and ROSAT (Maraschi et al. 1995a), and consistent with both within the 90% con?dence level. 3EG J1409-0745 (PKS 1406-076): This source was never studied in X-rays: ROSAT observation resulted only in an upper limit (2σ) with F0.1?2.4 keV < 2.35 × 10?13 erg cm?2 s?1 (Siebert et al. 1998). XMM-Newton observed this blazar twice, with the second observation about one month after the ?rst one, and the ?uxes in the ROSAT energy band were 4.2 and 3.9 × 10?13 erg cm?2 s?1 , respectively. In both observations, the source was best ?tted with a simple power law with Γ ≈ 1.6 and no additional absorption. 3EG J1621+8203 (NGC 6251): This EGRET source has been associated by Mukherjee et al. (2002) with the nearby FRI radio galaxy NGC 6251 (z = 0.02471). Later studies supported this conclusion, e.g. Sowards-Emmerd et al. (2003), Chiaberge et al. (2003), Guainazzi et al. (2003), Foschini et al.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 8

(2005). There is one XMM-Newton pointing available which lasted 50 ks, but heavy contamination with soft-proton ?ares strongly reduced the e?ective exposure on the three detectors (see Table A.1). The best ?t model is an absorbed broken power law model. The addition of a thermal plasma to the single power law does not provide an improvement with respect to the broken power law model. This data set has been analyzed by Gliozzi et al. (2004) and Sambruna et al. (2004), but there are some discrepancies with the present analysis (mainly the presence of the FeKα emission line) likely to be attributed to a di?erent cleaning of soft-proton ?ares. In the present analysis, the addition of a narrow (0.1 keV) or broad (0.5 keV) line at 6.4 keV, determines a worsening of the ?t both with respect to the single and the broken power law model. On the other hand, the present results are consistent with ASCA and BeppoSAX observations analyzed by Chiaberge et al. (2003) and Guainazzi et al. (2003) that found no indication of any emission line of the iron complex. An improvement with respect to the single power law is obtained by adding the mekal model, but the ?t remains always worse than the broken power law model. The parameters of the single power law plus thermal plasma model are: NH = (1.16 ± 0.08) × 1021 cm?2 , kT = 0.58+0.16 keV, ?0.13 Γ = 1.91 ± 0.03 for χ2 = 1.02 and 810 dof. The observed ?ux ? in the 0.4 ? 10 keV energy band is 5.81 × 10?12 erg cm?2 s?1 . 3EG J1830 ? 2110 (PKS 1830 ? 211): This is the highest redshift blazar in the present sample (z = 2.507) and is gravitationally lensed by an intervening galaxy at z = 0.886 (Wiklind & Combes 1996, Lidman et al. 1999, Courbin et al. 2002). It was observed in the past in X-ray by ROSAT (Mathur & Nair 1997), ASCA (Oshima et al. 2001) and Chandra (De Rosa et al. 2005). The present data set has not yet been published and two observations of the three available are a?ected by high background. 3EG J2158-3023 (PKS 2155-304): This BL Lac object is the third calibration source in the present sample and, for this reason, it has been observed several times. Most of the present data set have been already published (Edelson et al. 2001, Maraschi et al. 2002, Cagnoni et al., 2004, Zhang et al. 2005), where it is possible to ?nd more detailed analyses, particularly with reference to timing properties. This source, being one of the brightest in the X-ray sky, is also used in the search for the warm-hot intergalactic medium (WHIM) and a local absorber has been detected at 21.59 ? (≈ 0.57 keV) with the Re?ection Grating Spectrometers (RGS) on board XMMNewton (Cagnoni et al. 2004). However, since the study of this type of feature is outside the purpose of the present work, the spectrum of PKS 2155 ? 304 has been ?tted in the 0.6 ? 10 keV energy range and then the ?ux has been extrapolated to 0.4 keV.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 9

Table A.6. (left) Additional ?t results for Cen A (3EG J1324 ? 4314) with the model composed of an absorbed power law, plus one or two gaussian emission lines. Fit performed in the 4 ? 10 keV energy band and then extrapolated to 0.4 ? 10 keV. The two columns refer to the two observations. The rows list the following parameters: (1) Absorbing column density [1020 cm?2 ]; (2) Photon index of the power law model; (3) Normalization of the power law model [10?2 ph cm?2 s?1 keV?1 at 1 keV]; (4) Energy of the emission line 1 [keV]; (5) Line width σ 1 [keV]; (6) Flux of the emission line 1 [10?4 ph cm?2 s?1 ]; (7) Equivalent width of the emission line 1 [eV]; (8) Energy of the emission line 2 [keV]; (9) Line width σ 2 [keV]; (10) Flux of the emission line 2 [10?4 ph cm?2 s?1 ]; (11) Equivalent width of the emission line 2 [eV]; (12) Reduced χ2 and degrees of freedom; (13) observed ?ux in the 0.4 ? 10 keV energy band [10?11 erg cm?2 s?1 ]; (14) intrinsic luminosity in the 0.4 ? 10 keV energy band rest frame [1045 erg s?1 ]. The uncertainties in the parameter estimates are at the 90% con?dence limit for one parameter of interest. For NH = Gal, it means that the absorption column has been ?xed to the Galactic value. The luminosities were calculated using d = 3.84 Mpc. (right) EPIC spectrum in the 4 ? 10 keV energy band.
0093650201 1420+40 ?90 +0.07 2.24?0.04 +0.3 22.1?1.2 6.41 ± 0.02 < 0.074 +0.8 2.6?0.6 +22 70?16 +0.3 6.8?0.4 +0.39 0.76?0.25 +4.4 4.3?2.3 133+135 ?71 0.99/984 17.4 3.1 × 10?4 0093650301 1800 ± 200 2.3 ± 0.2 32+13 ?10 6.44 ± 0.02 < 0.065 +1.1 4.6?1.0 107+26 ?23 0.96/498 18.5 3.3 × 10?4

NH Γ Apl E1 σ1 AL1 EqW1 E2 σ2 AL2 EqW2 χ2 /dof ? F L

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 10

Table A.7. Optical properties of the AGN of the present catalog (from the Optical Monitor data). The data refers to the magnitude averaged over the whole observation. Columns: (1) Source name; (2) V magnitude [543 nm]; (3) B magnitude [450 nm]; (4) U magnitude [344 nm]; (5) UVW1 magnitude [291 nm]; (6) UVM2 magnitude [231 nm]; (7) UVW2 magnitude [212 nm]; The uncertainties in the parameter estimates are at the 1σ level and also include systematics.
Name (1) 0219 + 428 AO 0235 + 164 PKS 0521 ? 365 S5 0716 + 714 S5 0836 + 710 Mkn 421 V (2) B (3) U (4) UVW1 (5) 15.1 ± 0.1 17.0 ± 0.1 12.8 ± 0.1 16.1 ± 0.1 UVM2 (6) UVW2 (7)

13.4 ± 0.1 16.6 ± 0.1

12.9 ± 0.1 15.9 ± 0.1

12.8 ± 0.1 16.5 ± 0.1

PKS 1127 ? 145 ON 231 3C 273

15.7 ± 0.1 14.6 ± 0.1 12.6 ± 0.1 12.7 ± 0.1 12.7 ± 0.1 15.0 ± 0.1 12.9 ± 0.1 12.9 ± 0.1 12.9 ± 0.1 14.2 ± 0.1 11.8 ± 0.1 11.7 ± 0.1 11.8 ± 0.1 11.5 ± 0.1 11.5 ± 0.1 11.5 ± 0.1 11.3 ± 0.1

15.7 ± 0.1 11.3 ± 0.1 11.3 ± 0.1 11.3 ± 0.1 11.1 ± 0.1

15.9 ± 0.1 11.3 ± 0.1 11.3 ± 0.1 11.3 ± 0.1 11.1 ± 0.1

11.4 ± 0.1

11.3 ± 0.1

12.7 ± 0.1 Cen A? PKS 1334 ? 127 PKS 1406 ? 076 NGC 6251 PKS 1830 ? 211?

13.0 ± 0.1

11.8 ± 0.1

11.6 ± 0.1 16.3 ± 0.1 18.6 ± 0.1 19.4 ± 0.4 16.6 ± 0.1

11.5 ± 0.1 16.2 ± 0.1

11.4 ± 0.1

19.5 ± 0.2 15.6 ± 0.1 16.4 ± 0.1

19.7 ± 0.1

PKS 2155 ? 304

12.5 ± 0.1 12.1 ± 0.1 12.2 ± 0.1 13.2 ± 0.1 13.9 ± 0.1 13.8 ± 0.1 13.5 ± 0.1 14.3 ± 0.1 13.4 ± 0.1 14.2 ± 0.1 14.2 ± 0.1 12.4 ± 0.1 13.2 ± 0.1 12.9 ± 0.1 13.2 ± 0.1 13.1 ± 0.1 12.7 ± 0.1 12.7 ± 0.1 12.9 ± 0.1 12.6 ± 0.1 12.6 ± 0.1 12.8 ± 0.1 12.9 ± 0.1

12.6 ± 0.1

?

Source beyond the capabilities of OM: too faint or too bright.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 11

0219+428 z=0.444

0521-365 z=0.055

0716+714 z>0.3

0836+710 z=2.172

Mkn 421 z=0.033

1127-145 z=1.187

Fig. A.1. SED of the sources studied in this paper. We compare the XMM-Newton data (?lled black symbols) to the available archival data in all bands.

L. Foschini et al.: XMM–Newton observations of a sample of γ?ray loud active galactic nuclei, Online Material p 12

ON 231 z=0.102

3C 273 z=0.158

2155-304 z=0.117

NGC 6251

Cen A

Fig. A.2. SED of the sources studied in this paper. We compare the XMM-Newton data (?lled black symbols) to the available archival data in all bands.



更多相关文章:
...X-ray Jets in Radio-loud Active Galactic Nuclei.pdf
loud Active Galactic Nuclei_其它语言学习_外语学习_...ongoing observations of the Chandra X-ray ...Statistical studies for complete samples of blazars...
X-ray Luminosity Functions of Active Galactic Nuclei.pdf
X-ray Luminosity Functions of Active Galactic Nuclei_专业资料。In this ...cations. For the XMM-Newton Lockman Hole and Chandra DEEP ?eld samples, ...
The X-Ray Luminosity Function of Active Galactic Nuclei.pdf
The X-Ray Luminosity Function of Active Galactic Nuclei_专业资料。We derive...Recent BeppoSAX observations of a sample of low luminosity AGNs selected on...
...Multiwavelength Data of Radio-Loud Active Galactic Nuclei_....pdf
Data of Radio-Loud Active Galactic Nuclei_专业...Gamma Ray Observatory have made possible the ...For observations of a single source at di erent...
...of Seyfert galaxies from the Optical Monitor on XMM-Newton....pdf
exciting multi-wavelength dimension to observations of Active Galactic Nuclei. ...Symposium ‘New Visions of the X-ray Universe in the XMM-Newton and ...
...Sources suitable for NIR Adaptive Optics Observation.pdf
A Sample of X-Ray active extragalactic Sources ...X-ray surveys like ROSAT, Chandra, and XMM (e...Another source also shows a red nucleus (source ...
X-ray emission from active galactic nuclei with int....pdf
X-ray emission from active galactic nuclei with...spectral study of a sample of 8 AGNs with IM...We describe XMM-Newton observations and data ...
Feedback of Active Galactic Nuclei in Seyfert 2 Galaxies_论文....pdf
Feedback of Active Galactic Nuclei in Seyfert 2 Galaxies_天文/地理_自然科学_专业资料。ChnJAsrnAsrpy.18(08,.,5756i..to.thsVo.20)No534 o (t:wwc...
...XMM-Newton Survey of a Distance-Limited (D22 Mpc) Sample ....pdf
Key words: Galaxies: active - X-rays: galaxies...3. The XMM-Newton Sample Our team has been ...X-ray sources where (I) the nucleus dominates ...
Kinematics of AGN and Quasar Jets.pdf
of a complete sample of the the brightest 133 ...Recently TANAMI, (Tracking Active Galactic Nuclei ...(2005) found that gammaray blazars have more ...
The XMM-Newton Survey Science Centre Medium Sensitivity ....pdf
samples of carefully selected X-ray sources are ...elds. All XMMNewton observations, except for ...cation scheme: Broad-line Active Galactic Nuclei ...
...in Radio-Quiet Active Galactic.pdf
XMM-Newton observations of two luminous and high ...radio-quiet active galactic nuclei (AGNs) at z...X-ray spectra of a sample of nearby AGNs, ...
Spitzer Observations of Deeply Obscured Galactic Nuclei.pdf
Observations of Deeply Obscured Galactic Nuclei_专业...Active galaxies, with a favorable orientation of ...(2004) have analyzed a sample of ISO AGN ...
VLBI Surveys of Active Galactic Nuclei.pdf
VLBI Surveys of Active Galactic Nuclei_专业资料。A review is given on the...However, 22 and 43 GHz observations of the complete sample of compact ...
Synchrotron emission in the fast cooling regime which spectra....pdf
active galactic nuclei (AGNs), gammaray bursts...peak energy, in better agreement with observations...For example, if the plasma in the bulk where ...
...Very High Energy Gamma-Rays from Active Galactic Nuclei ....pdf
Gamma-Rays from Active Galactic Nuclei with the ...Observations of Mrk 421 in 2004 were carried out...[15] is applied to the final sample data to ...
...with High Resolution Chandra and XMM-Newton Spectra.pdf
Resolution Chandra and XMM-Newton Spectra_专业资料...X-ray binaries (XRBs), active galactic nuclei ...examples from observations of AGN and XRBs with ...
...XMM observations of the ELAIS-S1-5 sample_免费下....pdf
XMM observations of the ELAIS-S1-5 sampleAGN ...Aims. We use X-ray band observations to ...active galactic nuclei (AGN) and galaxies at ...
L00_intro张有宏讲义_图文.ppt
? 活动星系核 Active Galactic Nuclei (AGN) ? ...新星和伽马射线爆 (Supernovae and Gamma-ray ...Observations of Tycho Brahe 第谷 .布拉赫的观测 ...
X-ray properties of the Parkes sample of flat-spect....pdf
XMM-Newton observations ... 暂无评价 6页 免费...at-spectrum radio sources: dust in radio-loud ...ned samples of Active Galactic Nuclei (AGN), ...
更多相关标签:

All rights reserved Powered by 甜梦文库 9512.net

copyright ©right 2010-2021。
甜梦文库内容来自网络,如有侵犯请联系客服。zhit325@126.com|网站地图