The composite X-ray spectrum of 3CRR Quasars
DDraft version March 10, 2020
Typeset using L A TEX twocolumn style in AASTeX61
THE COMPOSITE X-RAY SPECTRUM OF 3CRR QUASARS
Minhua Zhou
1, 2 and Minfeng Gu Key Laboratory for Research in Galaxies and Cosmology, Shanghai Astronomical Observatory, Chinese Academy of Sciences, 80 NandanRoad, Shanghai 200030, China University of Chinese Academy of Sciences, 19A Yuquan Road, Beijing 100049, China
Submitted to ApJABSTRACTThe reason for the difference in the composite X-ray spectrum between radio-loud quasars (RLQs) and radio-quietquasars (RQQs) is still unclear. To study this difference, we built a new composite X-ray spectrum of RLQs by usingChandra X-ray data and Sloan Digital Sky Survey (SDSS) optical data for the sample of 3CRR quasars. We find theX-ray spectra of all 3CRR quasars except for 3C 351 have no soft X-ray excess and can be fitted with an absorbedpower-law model well. Our composite X-ray spectrum is similar to that of Shang et al. (2011) for RLQs, showinghigher hard X-ray and lower soft X-ray flux than the composite X-ray spectrum of RQQs. Most blazar-like 3CRRquasars have higher X-ray flux than the median composite X-ray spectrum, which could be related to the contributionof beamed jet emission at X-ray band. From the literature, we find that nineteen 3CRR quasars have extended X-rayemission related to radio jets, indicating inevitable contribution of jets at X-ray band. In contrast to RQQs, the X-rayphoton index of 3CRR quasars does not correlate with the Eddington ratio. Our results suggest that the jet emissionat X-ray band in RLQs could be related with the difference of composite X-ray spectrum between RLQs and RQQs.
Keywords: methods: statistical — catalogs — quasars: general — X-rays: general
Corresponding author: Minhua [email protected] author: Minfeng [email protected] a r X i v : . [ a s t r o - ph . H E ] M a r Zhou & Gu INTRODUCTIONX-ray emission appears to be nearly universal from ac-tive galactic nuclei (AGNs) and is often used to detectAGNs in various surveys (see e.g., Brandt & Hasinger2005; Watson 2012; Alexander et al. 2013; Brandt &Alexander 2015; Xue 2017). The intrinsic X-ray emis-sion from AGNs usually originates in the immediatevicinity of the supermassive black hole, and it consists ofseveral components including the primary X-ray emis-sion with a high energy cut-off, soft X-ray excess, reflec-tion and absorption components (e.g., Turner & Miller2009; Mallick et al. 2016). The primary X-ray emissionis usually thought to be from corona by inverse Comp-ton scattering of optical/UV photons (e.g., Haardt &Maraschi 1993). The origin of soft X-ray excess is stilldebated (Noda et al. 2013; Mallick et al. 2016; Jin et al.2017; Mallick, & Dewangan 2018), including the modelof thermal Comptonization in low temperature opticallythick medium that separated from primary X-ray emis-sion component (Magdziarz et al. 1998; Marshall et al.2003; Dewangan et al. 2007; Done et al. 2012), and themodel of blurred reflection from ionized accretion disk(Fabian et al. 2002; Crummy et al. 2006; Garc´ıa et al.2014).To systematically study the emission in quasars, Elviset al. (1994, hereafter E94) produced the first all-band(from radio to X-ray bands) spectral energy distribu-tions (SED) for a sample of quasars. Later, Shang etal. (2011, hereafter S11) built the next generation at-las SED of quasars. These SEDs are composite spec-tra constructed for quasar samples based on multi-bandobservational data. With the median values within fre-quency bins (e.g., S11), the composite spectrum can beregarded as the representative of overall SED (e.g., thebig blue bump, thermal IR bump, etc.) for the stud-ied sample, thus the systematic comparison on the SEDin different AGN populations can be readily performedby comparing their composite spectra. Both of thesetwo SEDs show prominent differences between radio-loud quasars (RLQs) and radio-quiet quasars (RQQs)at radio and X-ray bands, but have almost the samespectrum at other bands. While the difference in ra-dio band could be most likely due to the presence ofjets in RLQs, the reason for the difference in the X-rayband is still unclear. The possible reasons include theadditional jet-related UV/X-ray flux (e.g., Worrall etal. 1987; Miller et al. 2011), the more ionized accretiondisk (Ballantyne et al. 2002) or the X-ray emission fromthe hot Advection Dominated Accretion Flow (ADAF)within the truncated radius of accretion disk (Yuan, &Narayan 2014) in radio-loud AGNs. Many works found that radio-loud AGNs are moreX-ray luminous and usually have harder X-ray spectrathan radio-quiet AGNs (Zamorani et al. 1981; Worrallet al. 1987; Wilkes & Elvis 1987; Grandi et al. 2006;Kataoka et al. 2011; Miller et al. 2011; Wu et al. 2013;Zhu et al. 2019). As an early work, Worrall et al. (1987)found that the relative X-ray brightness is greater forRLQs and suggested that an “extra” X-ray emissionwould dominate the observed X-ray flux in the majorityof RLQs with flat radio spectra. Comparing with RQQs(Steffen et al. 2006), Miller et al. (2011) investigated the“excess” X-ray luminosity of RLQs relative to RQQs asa function of radio loudness and luminosity. They pro-posed that the X-ray emission of RLQs may consist ofdisk/corona and jet-linked components.Since RLQs usually have powerful jets, it could occurthe jet emission will contribute to X-ray spectra (Worrallet al. 1987; Miller et al. 2011), especially in flat-spectrumradio quasars (FSRQs). As a matter of fact, Grandi &Palumbo (2004) analyzed the X-ray spectrum of FSRQ3C 273 with BeppoSAX data and found a model with jetplus Seyfert-like components can fit the spectrum verywell. They found that the X-ray spectral index and softexcess flux have no correlation with total flux at X-rayband, but have a good correlation with the flux ratio ofthe jet to Seyfert components. It shows that the spectralindex tends to be flatter when the jet-like component hasmore contribution to the total flux.Since the jet is moving at small viewing angles inblazars (including FSRQs and BL Lac objects), the“beaming effect” is usually significant (Antonucci 1993;Urry & Padovani 1995). The jet emission in blazars canbe significantly boosted due to the “beaming effect”. Inother words, the observed emission can be much largerthan the intrinsic one caused by the Doppler effect fromrelativistic jet speed. This results in a nontrivial jetcontribution at all bands, like X-ray band studied inthis work, as indicated in 3C 273 above. Although theblazars were claimed to be excluded in both E94 andS11 work, we found 22 blazars out of 58 RLQs in S11sample, and 7 blazars out of 18 RLQs in E94 by check-ing with BZCAT catalog (Massaro et al. 2009, 2015)Edition 5.0.0 . In addition, the previous works on theX-ray difference between RLQs and RQQs may also in-clude blazars in their RLQs sample (e.g., Worrall et al.1987; Miller et al. 2011). In this work, we intend torevisit the difference of composite X-ray spectrum be-tween RLQs and RQQs, by building a new compositeX-ray spectrum for a sample of non-blazar RLQs. The -ray emissions of 3CRR quasars H = 70 km s − Mpc − , Ω m = 0 . Λ = 0 . A ( E ) = KE − Γ , where K is photons at 1 keV andE is photon energy. The spectral index α is defined as f ν ∝ ν − α with f ν being the flux density at frequency ν . SAMPLE3CRR catalog lists the brightest radio sources innorthern sky, and it was selected at low radio frequency178 MHz with flux density brighter than 10 . R varies from 0.5 to 4.3 ( R = f /f ˚ A ,Kellermann et al. 1989). The Chandra X-ray data isavailable for all quasars (Massaro et al. 2018; Stuardi etal. 2018). The majority of 3CRR quasars were detectedin SDSS (Abazajian et al. 2009; Alam et al. 2015) andTwo Micron All Sky Survey (2MASS, Cutri et al. 2003)photometric catalogs. Table 1 . The sample of 3CRR QuasarsName IAU name z Core flux log R Radio class Ref. CXO ID SDSS 2MASS(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)3C 9 0017+154 2.012 4.9 1.2 LDQ A05 1595 2000/11/30 y3C 14 0033+183 1.469 10.6 2.0 LDQ A05 9242 2008/11/02 y3C 43 0127+233 1.470 < . Table 1 continued
Zhou & Gu
Table 1 (continued)
Name IAU name z Core flux log R Radio class Ref. CXO ID SDSS 2MASS(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)3C 186 0740+380 1.063 15.0 1.9 CDQ L83 9774 2000/04/04 y3C 190 0758+143 1.197 73.0 2.9 LDQ A05 17107 2004/12/15 y3C 191 0802+103 1.952 42.0 2.5 LDQ A05 5626 2005/03/10 y3C 196 0809+483 0.871 7.0 1.5 LDQ A93 15001 2000/04/25 y3C 204 0833+654 1.112 26.9 2.0 LDQ A05 9248 2004/10/15 y3C 205 0835+580 1.534 20.0 1.5 LDQ A05 9249 2003/10/24 y3C 207 0838+133 0.684 510.0 3.5 LDQ A05 2130 2005/03/10 y3C 208 0850+140 1.109 51.0 2.3 LDQ A05 9250 2005/11/10 y3C 212 0855+143 1.049 150.0 3.2 LDQ A05 434 2005/12/06 y3C 215 0903+169 0.411 16.4 2.0 LDQ A05 3054 2005/01/19 y3C 216 0906+430 0.668 1050.0 4.1 CDQ L83 15002 2011/11/21 y3C 245 1040+123 1.029 910.0 3.6 LDQ A05 2136 2003/03/31 y3C 249.1 1100+772 0.311 71.0 1.7 LDQ A05 3986 2006/04/21 y3C 254 1111+408 0.734 19.0 1.8 LDQ A93 2209 2003/04/01 y3C 263 1137+660 0.652 157.0 2.1 LDQ A05 2126 2001/03/19 y3C 268.4 1206+439 1.402 50.0 2.2 LDQ A05 9325 2003/04/25 y3C 270.1 1218+339 1.519 190.0 3.1 LDQ A05 13906 2004/04/25 y3C 275.1 1241+166 0.557 130.0 3.0 LDQ A05 2096 2005/06/06 y3C 287 1328+254 1.055 2998.0 4.3 CDQ L83 3103 2004/12/21 y3C 286 1328+307 0.849 5554.0 4.3 CDQ L83 15006 2004/05/12 y3C 309.1 1458+718 0.904 2350.0 3.7 CDQ L83 3105 SSDC y3C 325 1549+628 0.860 2.4 1.8 LDQ This work 4818 2004/06/15 n3C 334 1618+177 0.555 111.0 2.4 LDQ A05 2097 2004/04/22 y3C 336 1622+238 0.927 20.4 2.2 LDQ A05 15008 2004/05/13 y3C 343 1634+628 0.988 < . < . Table 1 continued -ray emissions of 3CRR quasars Table 1 (continued)
Name IAU name z Core flux log R Radio class Ref. CXO ID SDSS 2MASS(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Note — Column (1): 3CRR name; Column (2): IAU name; Column (3): Redshift (Laing et al. 1983); Column (4): VLA5 GHz core flux in mJy with all from Laing et al. (1983), except for 3C 287 (Laurent-Muehleisen et al. 1997); Column (5):Radio loudness, R = f /f ˚ A , where f is rest-frame VLA 5 GHz core flux; Column (6): Radio classification,LDQ for Lobe dominated quasars, CDQ for Core dominated quasars; Column (7): References for radio classification, A05- Aars et al. (2005), A93 - Aldcroft et al. (1993), F03 - Fan & Zhang (2003), L83 - Laing et al. (1983), the classificationof 3C 325 is based on the core dominance value of 0.003 calculated from VLA measurements in Fernini et al. (1997);Column (8): Chandra observation ID; Column (9): The observational time of SDSS photometric data. When SDSS datais unavailable, the optical data were collected from NED or SSDC; Column (10): Available 2MASS data, y for yes , n for no . 3. DATA AND REDUCTIONTo create the composite X-ray spectrum of 3CRRquasars, we collected available optical/IR and X-raydata for all our sources.3.1.
Optical and Infrared data
There are 37 quasars detected by Sloan Digital SkySurvey (SDSS) at ugriz bands. For the remaining sixsources (3C 68.1, 3C 147, 3C 175, 3C 309.1, 4C 16.49and 3C 380), the photometric data at optical and/ornear infrared bands were collected from NASA/IPACExtragalactic Database (NED ) or Space Science DataCenter (SSDC ) (see Table 1).We obtained SDSS photometric data from SDSSDR12 , including fiber magnitude, fiber magnitude er-ror and extinction values at all bands. After correctingthe Galactic extinction, we converted SDSS magnitudes m AB to flux densities f ν using the zeropoint flux densityof f ν =3631 Jy (Oke & Gunn 1983).To produce the composite X-ray spectrum of 3CRRquasars, we followed the same method with S11, inwhich the rest-frame 4215 ˚A was used as a referencefor multi-band flux densities. In this case, we requiredthe available photometric data to cover the rest-frame4215 ˚A. In high-redshift quasars with 4215˚A not cov-ered by SDSS or NED/SSDC optical data, the near in-frared data were added (see details in Section 4.2). Weused 2MASS data archived by Infrared Science Archive(IRSA) for all quasars, except for 3C 270.1, in whichthe 2MASS data was taken from Krawczyk et al. (2013).The 2MASS magnitudes were converted to flux densitieswith zeropoint flux densities 1594, 1024, and 666.7 Jyfor J, H, and Ks bands, respectively (Cohen et al. 2003). http://ned.ipac.caltech.edu/ http://skyserver.sdss.org/dr12/en/home.aspx The near- to mid-IR data at 3.4, 4.6, 12 and 22 µ mare available for all 3CRR quasars from the WISE all-sky data release (Wright et al. 2010). In principle, thesedata can be used to build broadband optical to IR SEDsfor sample sources. However, we mainly focus on the fluxdensity at 4215˚A and the optical SED, thus the WISEdata were not used in this work.3.2. X-ray data ∼ (cid:48)(cid:48) ) has an incompa-rable advantage to isolate the core emission from theelongated jets. Thirdly, Chandra covers from 0.3 to 10keV band. It thus can be used to study the differenceat both soft and hard X-ray bands between RLQs andRQQs. For the sources with multiple observations fromChandra or SDSS, we chose those X-ray and optical datawith the closest time separation. However, when all thetime separation longer than one year, we selected X-raydata with longer exposure time. The selected X-ray andoptical data are shown in Table 1.The Chandra X-ray data of most 3CRR quasars havealready been analyzed in the literature (e.g., Crawford& Fabian 2003; Hardcastle et al. 2004; Croston et al.2005; Massaro et al. 2010, 2012, 2013, 2015). However,these works mostly focused on either the extended X-rayemission (e.g., Hardcastle et al. 2004) or only the X-rayflux (e.g., Massaro et al. 2015). The X-ray spectra ofmany 3CRR quasars have been studied with detailedspectral fitting in various works (Brunetti et al. 2002; Zhou & Gu
Aldcroft et al. 2003; Donahue et al. 2003; Fabian et al.2003; Belsole et al. 2006; Hardcastle et al. 2006; Siemigi-nowska et al. 2008; Hardcastle et al. 2009; Siemiginowskaet al. 2010; Wilkes et al. 2012, 2013). However, in theseworks, the spectra were analyzed by different groups andextracted from different regions likely causing system-atic difference between individual objects. Instead ofdirectly taking the results from the literature, we de-cided to re-analyze the selected Chandra data for 3CRRquasars in a uniform way.All 3CRR quasars were observed by Chandra X-rayObservatory (CXO) Advanced CCD Imaging Spectrom-eter (ACIS). We used Chandra Interactive Analysis ofObservations software (CIAO) v4.8 and Chandra Cal-ibration Database (CALDB) version 4.7.1 to reducethe data step by step with CIAO threads . We used chandra repro script to reprocess data and to createa new level 2 event file and a new bad pixel file. Af-ter that, we checked all sources for background flaresand filtered energy between 0 . −
10 keV. In the end,we extracted spectrum and response files from a source-centered circle of radius 2 . (cid:48)(cid:48) with specextract script.For the background region, we used 20 − (cid:48)(cid:48) annulus ifthere is no other sources in this region, otherwise, weused an arbitrary background region around the quasar.Before we analyzed the spectra, we carefully stud-ied the pile-up effect, i.e., when two or more photonsare detected as a single event. This effect will cause adistortion of the energy spectrum, such as two low en-ergy photons pile to a high energy photon. To confirmwhich source we should use pileup model in Sherpa, weused PIMMS to estimate the degree of pile-up. Af-ter inputting a series of parameters including the countrate of evt1 file in 1 . (cid:48)(cid:48) circle centered on source calcu-lated by F untools , a power-law spectrum with photonindex Γ = 2, redshift, Galactic H I Column Density (nH)(Dickey & Lockman 1990) and frame time of each obser-vation, PIMMS returns the pile-up fraction. As did inMassaro et al. (2012), we plotted the relation of pile-upfraction with evt1 counts per frame in Figure 1. A strongcorrelation was found between pile-up fraction and evt1counts per frame in 1 . (cid:48)(cid:48) circle. In this paper, we con-sider the pile-up effect according to the last version ofChandra Cycle 22 Proposers’ Observatory Guides .In the case that the pile-up effect can’t be ignored,there are two ways to extract and analyze the X-rayspectrum. One is to extract the spectrum from an an-nular area by excluding the piled-up core, for example, http://cxc.harvard.edu/ciao/threads/index.html http://cxc.harvard.edu/toolkit/pimms.jsp https://cxc.harvard.edu/proposer/POG/ P il e - u p f r a c t i o n [ % ] Figure 1.
Pile-up fraction versus evt1 counts per frame. . − . (cid:48)(cid:48) (e.g., Gambill et al. 2003; Worrall et al. 2004).The other is to fit the spectra extracted from the cen-tral region, which includes piled-up core, with jdppileup model (Davis 2001) in Sherpa. In this work, the lattermethod was adopted for those spectra extracted from2 . (cid:48)(cid:48) circle when the pile-up effect is significant, i.e., evt1count rate higher than 0 . / frame.We used Sherpa (Freeman et al. 2001) to fit all X-rayspectra for 3CRR quasars, and two statistical methodswere used (see Table 2). One is cstat for the case thatthe photon count from source-centered 2 . (cid:48)(cid:48) circle is lessthan 200, and thus the X-ray spectrum was basicallyunbinned. The other is chi xspecvar for the sourceswith enough exposure time, and the X-ray spectra werebinned with 15 counts per bin. RESULTS4.1.
X-ray spectra
The Chandra X-ray spectra extracted from source-centered 2 . (cid:48)(cid:48) circle were fitted with three models, in-cluding intrinsic absorbed power-law model with fixedGalactic absorption ( phabs ∗ zphabs ∗ powerlaw ), ab-sorbed broken power-law model ( phabs ∗ zphabs ∗ bknpower ) and absorbed double power-law model( phabs ∗ zphabs ∗ ( powerlaw powerlaw . -ray emissions of 3CRR quasars . − . . − . . − . conf script in Sherpa. The fluxand its uncertainties were calculated with sample f lux script in Sherpa.4.1.1. Compare with other works
There are various works dedicated to detailed spec-tral analysis on the Chandra data for some of 3CRRquasars with also simple power-law model. Wilkes etal. (2013) extracted X-ray photons from source-centered2.2 (cid:48)(cid:48) circle and presented the results of X-ray spectral fit-ting for nine 3CRR quasars (3C 9, 3C 186, 3C 191, 3C205, 3C 212, 3C 245, 3C 270.1, 3C 287, and 3C 432).We found that the X-ray photon index of their work isconsistent with our results within errors. The spectralanalysis on three 3CRR quasars (3C 47, 3C 215, 3C249.1) have been performed by Hardcastle et al. (2006).With extracted X-ray spectra from 1.25 (cid:48)(cid:48) circle, the re-sults of their work are generally in good agreement withour results. Belsole et al. (2006) and Gambill et al. (2003) studiedthe X-ray properties for nine quasars (3C 207, 3C 254,3C 263, 3C 275.1, 3C 309.1, 3C 334, 3C 345, 3C 380, and3C 454.3). From their results, the fitted X-ray spectraare in general flatter than our results. The difference islikely due to the PSF effect of Chandra/ACIS detectorthat the harder X-ray has broader PSF, since the centralpiled-up region was excluded to avoid pile-up effect intheir studies. This possibility is supported by our simu-lation, in which we found flatter X-ray spectra in outerregion than inner region (see Appendix B). The K − Stest shows significantly different distributions of X-rayphoton index in two regions. Using the same method toavoid the pile-up effect as Belsole et al. (2006), Hard-castle et al. (2009) presented the slightly flatter photonindex than our results for 3C 48, however the result of3C 325 is same as ours, in which there is no pile-upeffect. On the other hand, the similar X-ray photon in-dex with our results have also been found in the sourceswith pile-up effect, including 3C 48 in Siemiginowska etal. (2008), 3C 254 in Donahue et al. (2003), 3C 263 inHardcastle et al. (2002), and 3C 454.3 in Tavecchio etal. (2007).
Zhou & Gu T a b l e . T h e r e s u l t s o f X - r a y s p e c t r a l fi tt i n g . (cid:48)(cid:48) c i r c l e l o g f ( e r g c m − s − ) N a m e z n H C X O I D z . n H j dp . a j dp . f Γ N o r m . χ / d o f . S t a t . . − . k e V . − k e V . − . k e V ( )( )( )( )( )( )( )( )( )( )( )( )( )( ) C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E + . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . T a b l e c o n t i n u e d -ray emissions of 3CRR quasars T a b l e ( c o n t i n u e d ) . (cid:48)(cid:48) c i r c l e l o g f ( e r g c m − s − ) N a m e z n H C X O I D z . n H j dp . a j dp . f Γ N o r m . χ / d o f . S t a t . . − . k e V . − k e V . − . k e V ( )( )( )( )( )( )( )( )( )( )( )( )( )( ) C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . C . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E + . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . . E − . . . + . − . . + . − . . c h i − . + . − . − . + . − . − . + . − . C . . . E − . + . − . . + . − . . c s t a t − . + . − . − . + . − . − . + . − . N o t e — C o l u m n ( ) : C RR n a m e ; C o l u m n ( ) : R e d s h i f t ; C o l u m n ( ) : G a l a c t i c n e u tr a l H y d r og e n c o l u m nd e n s i t y ( K a l b e r l a e t a l. ) ,i nun i t s o f c m − ; C o l u m n ( ) : C h a nd r ao b s e r v a t i o n I D ; C o l u m n ( ) : I n tr i n s i c H y d r og e n c o l u m nd e n s i t y ,i nun i t s o f c m − ; C o l u m n s ( - ) : T h e a l ph aa nd f p a r a m e t e r s i n S h e r p a f o r j d p i l e u p m o d e l ( D a v i s ) ; C o l u m n ( ) : T h e p o w e r - l a w ph o t o n i nd e x a nd σ e rr o r s ; C o l u m n ( ) : T h e n o r m a li z a t i o n a nd σ e rr o r s o f p o w e r - l a w c o m p o n e n t i n − ph o t o n s k e V − c m − s − a t k e V ; C o l u m n ( ) : R e du ce d χ a ndd e g r ee o ff r ee d o m ; C o l u m n ( ) : T h e s t a t i s t i c a l m e t h o d , c h i f o r c h i x s p e c v a r , c s t a t f o r c s t a t ; C o l u m n s ( − ) : T h e fl u x e s a nd σ e rr o r s a f t e r a b s o r p t i o n c o rr ec t i o n s . Zhou & Gu spec.pi C oun t s / s e c / k e V Energy (keV) S i g m a −4−202 Figure 2.
An example of X-ray spectral fitting. A singlepower-law can fit the spectrum well in 3C 275.1.
Composite spectra
As shown in S11, the composite spectrum constructedfrom the median values within frequency bins (e.g., S11)can be used as the representative of the overall SEDfor the studied sample. The median values at binnedfrequencies were obtained from the SED normalized toa given optical frequency. For this reason, the opti-cal/infrared SED will be needed in order to estimatethe normalized optical flux. Therefore, to constructthe composite X-ray spectrum for the sample, we firstlybuilt the log( νf ν ) − log ν SED at optical/infrared andX-ray bands of all quasars by directly calculating therest-frame flux density f ν obs (1+ z ) from observational op-tical/infrared data f ν obs in combination with X-ray mea-surements. As examples, the SEDs of four quasars areshown in Figure 3.We followed the same method in S11 to construct thecomposite spectrum. The SED of individual objects wasfirstly normalized to rest-frame 4215 ˚A. The flux den-sity at rest-frame 4215 ˚A was directly estimated from thepower-law fit on the optical/infrared data. When 4215 ˚Ais covered by SDSS/NED/SSDC data, only these opticaldata were used in power-law fit. For those quasars at0 . < z ≤ .
2, the 2MASS J-band data was added in thefit. For the objects at 1 . < z ≤ .
0, 2MASS J,H bandswere added, and 2MASS J,H,K bands were used when z > .
0. In all the cases except for 3C 68.1, the power-law model gave good fit to the continuum. The contin- uum of 3C 68.1 is convex, thus deviated from power-law(see Figure 3). This is likely caused by heavy extinctionas shown in Appendix A.2. Therefore, we fitted the con-tinuum with the log-parabolic model ( f ν = kν ( a + b log ν ) ,Massaro et al. 2004).After normalization, the optical/IR continuum was re-binned with 12 bins in the log ν range of 14.7 to 15.3Hz. This frequency range was selected as it is coveredby most quasars. At X-ray band, the spectra were re-sampled in the log ν range of 17 . − .
45 Hz, with a binsize of log ν = 0 .
1. Following S11, the composite SEDat optical/IR and X-ray band was constructed from themedian values in each bin. Two composite median spec-tra were finally constructed, with one for all 43 3CRRquasars, and the other after excluding six blazars. Thecomposite spectra are shown in Figure 4 with red lines.it can be clearly seen that the composite X-ray spectrumof our 3CRR quasars is very close to the composite spec-trum of RLQs in S11, however, it differs that of RQQsin S11 with flatter and stronger hard X-ray emission. DISCUSSIONS5.1.
Selection effect
There is a well-known relationship between α ox and L ˚ A (e.g., Lusso et al. 2010), in the way that moreluminous sources have steeper optical to X-ray slope,which is defined as α ox = − .
384 log( L /L ˚ A ) . (1)In this work, the composite X-ray spectrum is con-structed using rest-frame 4215 ˚A as a reference, there-fore, it will largely depend on the optical/UV luminosity.When comparing with other samples, like S11, this de-pendence needs to be investigated in order to avoid anyselection effects.In Figure 5, we plotted the relationship between α ox and L ˚ A for our sample and S11 sample. The lumi-nosity at 2500 ˚A of S11 objects is taken from Tang et al.(2012). As Figure 5 shows, 3CRR quasars and S11 RLQshave similar distribution on α ox − L ˚ A panel. How-ever, S11 RQQs have larger α ox (mostly > .
2) than thatof S11 RLQs and 3CRR quasars. While the L ˚ A lumi-nosity of 3CRR quasars are similar to those of S11 RLQswith Kolmogorov Smirnov (K − S) test values D = 0 . P = 0 . D = 0 .
632 and P = 9 . E −
06 from K − Stest.The similar composite X-ray spectrum of our sam-ple with S11 RLQs seems to be reasonable consideringtheir similar optical/UV luminosity. In contrast, thehigher L ˚ A for both samples than S11 RQQs, would -ray emissions of 3CRR quasars
14 15 16 17 18 19Rest-frame log( ) [Hz]1816141210 l o g ( f ) [ e r g s c m ]
3C 351 14 15 16 17 18 19Rest-frame log( ) [Hz]1816141210 l o g ( f ) [ e r g s c m ]
3C 20514 15 16 17 18 19Rest-frame log( ) [Hz]1816141210 l o g ( f ) [ e r g s c m ]
3C 9 14 15 16 17 18 19Rest-frame log( ) [Hz]1816141210 l o g ( f ) [ e r g s c m ]
3C 68.1
Figure 3.
The optical/IR and X-ray SED of four quasars, shown as examples for our sample. The power-law fits are shown inthe plots, except for the log parabolic model for optical/IR data of 3C 68.1 (see text for details). The dashed line representsthe position of normalization wavelength 4215 ˚A. imply lower X-ray emission if they follow the general α ox − L ˚ A relation (e.g., Green et al. 1995; Ander-son et al. 2003; Vignali et al. 2003). However, this isopposite to our finding, i.e., higher X-ray emission inour sample compared to S11 RQQs. Therefore, the dif-ference between RLQs and RQQs cannot be driven byselection effect related with optical/UV luminosity.In addition to optical/UV luminosity, we comparedblack hole mass M BH and Eddington ratio L bol /L Edd of 3CRR quasars with S11 RLQs and RQQs, as shownin Figure 6. The black hole mass and Eddington ra-tio of S11 quasars were directly taken from Tang et al.(2012). The black hole masses of all 3CRR quasars wereobtained from McLure et al. (2006), except for 3C 216and 3C 345 (Shen et al. 2011), 3C 343 and 3C 455 (Wu2009), and 3C 454.3 (Gu et al. 2001). All these virialblack hole masses were estimated with the empirical re-lationship between the broad line region radius and theoptical/UV continuum luminosity, in combination withthe line width of broad emission lines (e.g., Shen et al.2011). Depending on source redshift and availability of emission lines, various lines were used in the literaturefor our sample sources, with H β usually at low-redshiftsources, while Mg II or C IV at high redshift (see e.g.,Shen et al. 2011). We estimated the bolometric luminos-ity L bol of 3CRR quasars with the relation establishedfrom S11 sample in Runnoe et al. (2012),log( L iso ) = (4 . ± .
66) + (0 . ± .
04) log( λL λ, ˚ A )(2)The Eddington luminosity was calculated with blackhole mass as L Edd = 1 . × ( M BH /M (cid:12) ) erg s − (Tang et al. 2012).We found from Figure 6 that the Eddington ratio of3CRR quasars, S11 RLQs and RQQs are similar, al-though the black hole masses of 3CRR quasars and S11RLQs are systematically larger than those of S11 RQQs.This implies a similar accretion mode in all three sam-ples. Thus, it further indicates that the difference incomposite X-ray spectrum between RLQs and RQQs isnot caused by selecting different accretion systems, asmanifested from the dependence of α ox on the Edding-ton ratio (e.g., Ruan et al. 2019).2 Zhou & Gu
15 16 17 18Rest-frame log( ) [Hz]32101 l o g ( f ) [ A r b i t r a r y U n i t s ]
15 16 17 18Rest-frame log( ) [Hz]32101 l o g ( f ) [ A r b i t r a r y U n i t s ] Figure 4.
The median composite SED at optical and X-ray bands normalized at 4215˚A for our sample of 3CRR quasars, whichare shown in red solid lines. The gray solid lines are normalized SED of individual objects. The thick black and blue solid linesare composite SEDs for RLQs and RQQs in Shang et al. (2011), respectively. The green solid lines are composite X-ray spectraafter extinction correction (see text for details).
Upper − for all 3CRR quasars, and the orange solid lines are X-ray spectra forsix blazars; Bottom − for 3CRR quasars excluding six blazars. -ray emissions of 3CRR quasars L , 2500Å ) [erg s Hz ]0.81.01.21.41.61.82.02.2 O X N Figure 5.
Relation between α ox and luminosity L ˚ A for3CRR and S11 quasars. The upper panel shows the his-togram of luminosity L ˚ A . M BH ) [M ]2.01.51.00.50.00.5 l o g ( L b o l / L E dd ) Figure 6.
The distribution of black hole mass and Edding-ton ratio for 3CRR quasars and S11 samples. The right panelshows the histogram of Eddington ratio.
Extinction
The advantage of selecting sources at radio band isthat the radio emission is not subject to dust extinction,which is in contrast to optical selected sample. We foundthat the composite optical spectrum of 3CRR quasarsis redder than S11 one (see Figure 4). Indeed, we foundsteep/red optical/UV spectra in many sources (e.g., 3C14, 3C 68.1, 3C 190, 3C 205, 3C 212, 3C 216, 3C 268.4,3C 270.1, 3C 325, 3C 343, 3C 345, 3C 454.3 and 3C 455), which can be seen in Figure 4. Prominent absorp-tion lines were found in the optical spectra of many ofthese quasars, such as Si II , Mg II , C IV , Fe II etc. (e.g.,Aldcroft et al. 1994). It seems that there is some ab-sorber locating at the line between source and observer.As the most extreme case, 3C 68.1 is likely a highlyinclined and reddened quasar (Boksenberg et al. 1976;Brotherton et al. 1998) as mentioned earlier. 3C 325 isalso a reddened quasar classified by Grimes et al. (2005),and its spectrum shows lots of absorption lines in blueside. 3C 270.1 has a steep extreme ultraviolet (EUV)spectrum (Punsly & Marziani 2015). n H [ c m ] Figure 7. H I column density versus optical spectral index α ν . The dashed line indicates the spectral index of the com-posite spectrum of S11 RLQs ( α ν = 0 . In Figure 7, we plot the relation of intrinsic H I columndensity (nH, see Table 2) and optical spectral index ( α ν )of 3CRR quasars, where nH was calculated from X-rayspectral fitting. We found that the quasars with redderspectra tend to have relatively larger H I column density,supporting dust extinction in some of these objects. Ifthe emission at 4215 ˚A is indeed dust extincted, then theX-ray emission will be overestimated in the compositespectrum.The systematic extinction of our 3CRR quasar samplecan be evaluated by constructing the extinction curveby comparing our composite optical/UV spectrum withthat of S11 RLQs sample. We normalized two compos-ite spectra at log ν = 14 .
725 Hz (i.e., λ = 5647 . Zhou & Gu l o g ( f ) [ A r b i t r a r y U n i t s ] [ m ]505101520 E ( V ) / E ( B V ) GALAXY ( R v = 3.1)3CRRGaskell et al. (2004)SMC Figure 8.
Extinction in 3CRR quasars: ( upper ) - The com-posite optical/UV spectrum of 3CRR quasars (black crosses)and S11 RLQs (black solid line). The dotted line is thepower-law fit on the line-free windows (black triangles) on theS11 RLQs spectrum; ( bottom ) - Extinction curve of 3CRRquasars shown with black crosses. The filled triangles arefrom the reddening curve of 0 . ≤ R c < R c is radio core-to-lobe flux ratio. Thesolid line is Galactic reddening curve of R V = 3 .
1. The filledcircles show the extinction curve of Small Magellanic Cloud(SMC, Prevot et al. 1984). ( R V = 3 .
1) (Cardelli et al. 1989), SMC extinction curve,and the reddening curve of lobe-dominant radio AGNsin Gaskell et al. (2004), with significantly higher valueat short wavelength.The extinction at 4215 ˚A, A , is tentatively esti-mated by assuming a selective extinction R V = 3 .
1. Inthis case, we get A = 0 . R V = 5 . Soft X-ray excess
Compared to S11 RQQs, both composite X-ray spec-tra of our sample and S11 RLQs show lower soft X-rayemission. This difference may possibly be caused by thelow fraction of soft X-ray excess detected in RLQs com-pared to RQQs (Scott et al. 2011, 2012; Boissay et al.2016).There are 27 RQQs in S11, of which 23 sources haveavailable X-ray spectra, 7 from ROSAT and 16 fromChandra or XMM-Newton observations. As shown fromthe spectra fitting in S11, we found that all ROSAT datawere fitted by a single power-law with a steep photonindex (Γ > . X-ray emission in RLQs
It’s well known that the difference of radio emissionon the composite SED between RLQs and RQQs (Elviset al. 1994, S11) can be explained by the presence ofpowerful relativistic jets in RLQs, which is either weakor absent in RQQs (Panessa et al. 2019). It is conceiv-able that the jet emission may contribute also at otherbands than radio band (e.g., Blandford et al. 2019), es-pecially in blazars when the relativistic jets is moving atsmall viewing angles (Antonucci 1993; Urry & Padovani1995). While the X-ray emission in RQQs may mainlybe from the disk-corona system, the additional contribu-tion from jets may present in RLQs (e.g., Worrall et al.1987; Miller et al. 2011; Li 2019). However, the fractionof X-ray emission in radio-loud AGNs that is from the -ray emissions of 3CRR quasars
Comparison with previous works
In Figure 4, while our composite X-ray spectrum issimilar to that of S11 RLQs, we found that almost allsix blazars (3C 207, 3C 216, 3C 309.1, 3C 380, 3C 345,and 3C 454.3) in our sample have higher X-ray flux thanthe composite spectrum, which may be caused by thesignificant jet contribution as the jet emission is usu-ally thought to be highly boosted due to “beaming ef-fect”. However, although the X-ray composite spectrumof 3CRR quasars after excluding six blazars is lower thanthe original one, it is still close to that of S11 RLQs (seeFigure 4). The student’s t -test shows a significant differ-ence at >
99% confidence level between the distributionsof normalized X-ray flux at log ν = 18 .
22 Hz of 3CRRquasars and S11 RQQs. As a result, the difference ofcomposite X-ray spectrum between RLQs and RQQs isstill significant.The more luminous X-ray in radio-loud AGNs com-pared to the radio-quiet population, has been also foundin many works (e.g., Zamorani et al. 1981; Worrall et al.1987; Miller et al. 2011). Zamorani et al. (1981) foundthat RLQs have more X-ray luminosity than RQQs atgiven optical luminosity. Worrall et al. (1987) suggestedthat an “extra” X-ray emission would dominate the ob-served X-ray flux in the majority of RLQs with flat radiospectra. Miller et al. (2011) combined large, modern op-tical (e.g., SDSS) and radio (e.g., FIRST) surveys witharchival X-ray data from Chandra, XMM-Newton, andROSAT to generate an optically selected sample that in-cludes 188 RIQs and 603 RLQs. The authors found thatthe excess X-ray luminosity compared to RQQs rangesfrom ∼ . ∼ >
10 for strongly radio-loud or luminous objects. Based on the results, theyproposed a model in which the nuclear X-ray emissioncontains both disk/corona-linked and jet-linked compo-nents and demonstrated that the X-ray jet-linked emis-sion is likely beamed but to a lesser degree than appliesto the radio jet.On the other hand, Wilkes & Elvis (1987) found thatthe X-ray spectral slope of RLQs is flatter than that ofRQQs, and argued that the “two-components” model(the flat and the steep components dominate the X-ray flux in RLQs and RQQs, respectively) can explain theresult. On the contrary, Sambruna et al. (1999) providedevidence that radio-loud AGNs show comparable distri-bution of the X-ray continuum slope with radio-quietAGNs. Similarly, Gupta et al. (2018) found that radiogalaxies are on average X-ray-louder than radio-quietAGNs, however, the spectral slopes of two populationsare very similar. The authors argued that in radio-loudand radio-quiet AGNs the hard X-rays are producedin the same region and by the same mechanism. Thelarger X-ray luminosities in radio-loud AGNs may re-sult from larger radiative efficiencies of the innermostportions of the accretion flows around faster rotatingblack holes. More recently, Gupta et al. (2020) foundthat the average X-ray loudness of Type 1 and Type2 radio-loud AGNs is very similar based on the sampleselected from the
Swif t /BAT catalog. The X-ray loud-ness defined as the ratio of hard X-ray luminosity at14-195 keV to MIR luminosity can be used to study theorientation-dependent X-ray emission since the MIR ra-diation is expected to be isotropic (see e.g., Lusso et al.2013). This similarity indicates negligible dependenceof the observed X-ray luminosities on the inclinationangle. Therefore, this seems to disfavor the significantjet emission at X-ray band as otherwise the jet emis-sion will be expected to be more significant at smallerviewing angle, then higher X-ray loudness in Type 1sources. As found in 3C 273, the well-known blazar,the model-fit on the X-ray spectrum from combined ob-servations with Chandra, INTEGRAL, Suzaku, Swift,XMM-Newton, and NuSTAR observatories, gives thatthe coronal component is fit by Γ
AGN = 1 . ± . E cutoff = 47 ±
15 keV, and jet photon in-dex by Γ jet = 1 . ± . −
40 keV.The photon index at 2 −
10 keV in our 3CRR quasarsranges from 1.45 to 2.32 with a median value of 1.69.This is in comparable with RQQs (e.g., Wang et al.2004; Brandt & Alexander 2015), in good agreementwith Gupta et al. (2018) that radio-loud and radio-quietAGNs have similar distribution of hard X-ray photonindex. A significant correlation between the X-ray pho-ton index and the Eddington ratio has been found inradio-quiet AGNs in various works, in which the X-rayemission is thought to be from disk-corona system (e.g.,Wang et al. 2004; Brandt & Alexander 2015). Such cor-relation if found would be a clue on the emission mech-anism of X-ray emission. We plot the relation of theX-ray photon index with the Eddington ratio for our3CRR quasars sample in Figure 9. There is no signifi-cant correlation between the X-ray photon index and the6
Zhou & Gu
Eddington ratio, in contrast to the finding for RQQs inWang et al. (2004); Brandt & Alexander (2015). A sim-ilar result has also been recently found in Li (2019) fora well-selected sample of radio-loud AGNs. This seemsto imply that the X-ray emission may not be from, orat least not be dominated by the disk-corona system.In this case, it seems that the jet contribution can’tbe ignored. As a matter of fact, the photon index at2 −
10 keV of our sample is similar to that of FSRQs,which have an average value of 1 . ± .
04 (Donato et al.2001). However, the large uncertainties in the photonindex, probably caused by low data quality, preclude usto draw a firm conclusion. L bol / L Edd )1.52.02.53.0
Figure 9.
The relation between the X-ray photon index at2 −
10 keV and the Eddington ratio.
X-ray jets
In principle, the decomposition of X-ray emission intodisk-corona and jet component can be performed when ahigh-quality X-ray spectrum is available at broad bands,as did in 3C 273 (Madsen et al. 2015). However, this isusually hard for sample study due to the lack of broadband data. Although the fraction of X-ray emission inradio-loud AGNs that is from the jet is still unclear,it’s contribution can be studied from the morphology ofX-ray emission. Besides the unresolved central X-raycomponent, the X-ray emission can also be seen fromthe outer jet as shown in Harris & Krawczynski (2006).The X-ray emission associated with radio jet knots hasbeen studied in details for radio sources in 3CR catalog(Massaro et al. 2015).XJET (Harris et al. 2010; Massaro et al. 2010,2011a,b) collected 117 sources, which have extendedX-ray emission associated with radio jets. After cross- matching 3CRR quasars with XJET catalog, we found11 quasars have extended jet component in X-ray im-ages, which has been studied in the literature, including3C 9 (Fabian et al. 2003), 3C47 (Hardcastle et al. 2004),3C 207 (Brunetti et al. 2002), 3C 212 (Aldcroft et al.2003), 3C 254 (Donahue et al. 2003), 3C 263 (Hardcas-tle et al. 2002), 3C 275.1 (Crawford & Fabian 2003),3C 345 (Sambruna et al. 2004), 3C 351 (Brunetti et al.2001), 3C 380 (Marshall et al. 2005) and 3C 454.3 (Mar-shall et al. 2005). Moreover, we checked the results ofMassaro et al. (2015), and found that six more quasars(3C 181, 3C 191, 3C 215, 3C 245, 3C 325 and 3C 334)may have extended X-ray component related with jetknot, hotspot, or lobe. In addition, the extended X-ray emission from jet has also been found in 3C 270.1(Wilkes et al. 2012) and 3C 432 (Erlund et al. 2006).In total, there are 19 3CRR quasars having extendedX-ray components associated with radio jets.Massaro et al. (2009) reported the X-ray emission fromthe radio jet of 3C 17 with Chandra observations, andfound that the high energy emission from the jet knotscan be explained by both inverse Compton(IC)/cosmicmicrowave background (CMB) and Synchrotron process.Recently, Mingo et al. (2017) found that the spectral fitof hotspots is consistent with the synchrotron emission,while the IC emission is found for lobes. In the casestudy of 3C 459, the extended X-ray emission can bewell modelled by a plasma collisionally heated by jet-driven shocks (Maselli et al. 2018). Based on the goodcorrelation between the unabsorbed component of X-rayluminosity and the 5-GHz core radio luminosity, Hard-castle et al. (2009) argued that at least some, and inmany cases all, of the soft component of radio-source X-ray spectra originates in the jet is very hard to evade fora sample of 3CRR radio sources. As shown in Massaroet al. (2015), jet knot, hotspot, and lobe may all haveX-ray emission, therefore, the emission from the jet com-ponents in the region of 2 . (cid:48)(cid:48) radius will contribute in thecomposite spectrum constructed for our quasar sample.While the central 2 . (cid:48)(cid:48) is unresolved in X-ray image, itcan be usually resolved into core-jet structure with acentral bright core and several jet components at radioband (e.g., 3C 245, see Figure 1 in Massaro et al. 2015).When the X-ray emission is extracted in central 2 . (cid:48)(cid:48) ,the X-ray emission associated with jet components willnaturally be included, although their flux fractions arelargely unknown. As an additional check of jet emissionat X-ray band, we studied the X-ay emission at differentregions. However, no strong evidence of jet contributionwas found due to the large uncertainties in the X-rayphoton index (see Appendix B). -ray emissions of 3CRR quasars SUMMARYIn this work, we revisited the difference of compos-ite X-ray spectrum between RLQs and RQQs, by usingmulti-band data for 3CRR quasars sample, which wasselected at low radio frequency, thus less affected by the“beaming effect”. We found that the composite X-rayspectrum of 3CRR quasars is similar to S11 RLQs af-ter excluding blazars, still significantly different fromRQQs. Although the photon index at 2 −
10 keV of oursample is similar to RQQs, there is no strong correlationbetween the photon index and Eddington ratio, imply-ing that other emission than the one from disk-coronasystem may also contribute at X-ray band, presumablyfrom jet. The detection of X-ray emission from jet com-ponents has been reported in many sources from theliterature, and the jet components might be included inextracting the X-ray spectrum from the central 2 . (cid:48)(cid:48) Facilities:
SDSS, 2MASS, CXO
Software:
CIAO (Fruscione et al. 2006), HEASoft ,PythonAPPENDIX A. THE X-RAY SPECTRAL FITA.1.
3C 351
3C 351 ( z = 0 . . (cid:48)(cid:48) region are 0.142 and 0 .
327 counts / frame, for 435 and 2128 data, respectively.Therefore, the pile-up effect might be taken into account for the spectrum of 2128. We fitted these two spectra withabsorbed power-law model and found that both spectra cannot be fitted well with significant excess at soft X-ray band(see the left panel in Figure 10).The soft X-ray spectrum of 3C 351 has been investigated by Fiore et al. (1993) using ROSAT data. The authorsfound that partial covering, soft excess, and warm absorber models can all fit the spectrum well, although only thewarm absorber model (i.e., partially ionized absorbing material in the line of sight) gives a good χ for a typical valueof high energy continuum slope. If the warm absorber model is correct, the strongest absorption edge lies in the range0 . − .
76 keV, implying O IV − O VII as the most likely absorbing ions. Mathur et al. (1994) provided strong evidencefor the warm absorber model in the soft X-ray spectrum.As did in Fiore et al. (1993), we fitted the spectra of 3C 351 with the warm absorbed power-law model ( phabs ∗ zxipcf ∗ powerlaw ) using chi xspecvar statistical method. Here, we set the nH of photo-electric absorption ( phabs )to Galactic H I column density (Kalberla et al. 2005) and zxipcf redshift to 0.371. We found two Chandra spectra8 Zhou & Gu spec.pi C oun t s / s e c / k e V Energy (keV) S i g m a −6−4−20246 spec.pi C oun t s / s e c / k e V Energy (keV) S i g m a −202 Figure 10.
3C 351 spectrum of Chandra observation ID 435: (
Left ) - The spectrum is fitted by the absorbed power-law model( phabs ∗ zphabs ∗ powerlaw ); ( Right ) - The spectrum is fitted with the warm absorbed power-law model ( phabs ∗ zxipcf ∗ powerlaw ). Table 3.
The spectral fitting for 3C 351
435 2128(1) (2) (3) (4)pile-up alpha 0.42f 1.00zxipcf nH(10 cm − ) 3.00 2.96log( xi ) 1.90 1.20Covering Fraction 1.00 0.90Power-law Γ 1 . +0 . − . . +0 . − . Norm. (10 − ) 4 . +7 . − . . +0 . − . χ Red /dof. . /
114 1 . / . − . − . +0 . − . − . +0 . − . . − . − . +0 . − . − . +0 . − . . − . − . +0 . − . − . +0 . − . Note —Rows 1 and 2: pile-up parameters; Rows 3 − zxipcf parameters; Rows 6 −
8: power-law parameters, where Γ is pho-ton index, Norm. is photons keV − cm − s − at 1 keV, and χ Red /dof. is reduced χ and degree of freedom; Rows 9 − f in units of erg cm − s − ). can be well fitted by the model. As shown in the right panel of Figure 10, the prominent absorption edge is visiblebelow 1 keV. The results of spectral fitting are given in Table 3. -ray emissions of 3CRR quasars
3C 68.1
3C 68.1 is an optically red quasar with very steep optical spectrum ( α ∼ .
1) (Boksenberg et al. 1976), and is alsohighly polarized at 3000 − II , and Ca II K. 3C 68.1 probablyhas absorbed X-ray emission (Bregman et al. 1985). All these observational features indicate that 3C 68.1 is likely anedge-on system along a line of sight through dusty, and ionized gas, perhaps part of torus (Brotherton et al. 2002).3C 68.1 has been observed with an exposure time of 3 .
05 ks by Chandra on February 10, 2008 (ID: 9244). TheX-ray spectrum extracted from 2 . (cid:48)(cid:48) circle was fitted with an absorbed power-law model with fixed Galactic absorption( phabs ∗ zphabs ∗ powerlaw ). An extremely flat spectrum was found with Γ = 0 . +0 . − . . The flat X-ray spectrum isconsistent with the finding in Goulding et al. (2018) for extremely red quasars, and it may imply severe absorption inthe source as shown from the extremely red optical spectrum and intrinsic ultraviolet absorption. Therefore, we fittedthe X-ray spectrum again by fixing Γ = 1 .
9, and found the intrinsic X-ray absorption of N H ∼ . × cm − . B. X-RAY EMISSION AT DIFFERENT REGIONSTo further get the clues of jet contribution at X-ray band, we tried to compare the X-ray spectra extracted fromdifferent regions, 0 . − . (cid:48)(cid:48) annulus and central 1 . (cid:48)(cid:48) , in which the jet contribution is expected to be different. 3C 68.1and 3C 343 are excluded due to low photon counts in 0 . − . (cid:48)(cid:48) annulus. We found that all Chandra X-ray spectracan be well fitted by absorbed power-law model except for 3C 351 (see Table 4).The distribution of X-ray photon index at 2 −
10 keV from 0 . − . (cid:48)(cid:48) , 1 . (cid:48)(cid:48) and 2 . (cid:48)(cid:48) are compared in Figure 11.While the photon index of 1 . (cid:48)(cid:48) is similar to that of 2 . (cid:48)(cid:48) , we found that most sources (34/41) have flatter spectra fromouter region (0 . − . (cid:48)(cid:48) annulus) than inner region (1 . (cid:48)(cid:48) circle). While this may be caused by the intrinsic differencein the emission at different regions, the effect that Chandra/ACIS detector has broader PSF at hard X-ray can’t beignored. To examine the PSF effect, we used MARX simulate events on the X-ray data of our sample. After excludingthe sources with pile-up effect, and low photon counts data, we extracted the X-ray spectra of 26 quasars from thesource-centered region with 5 (cid:48)(cid:48) radius out of background region (5 − (cid:48)(cid:48) annulus) and fitted them with 1-D polynomialfunction ( polynom d ), which were used as input spectra to MARX. The X-ray spectra of source-centered 1 . (cid:48)(cid:48) circleand 0 . − . (cid:48)(cid:48) annulus extracted from simulated data were fitted with an absorbed power-law model ( phabs ∗ powerlaw ).We found the X-ray spectra from outer region are flatter than those from inner region, with median photon index of1.49 and 1.68 for the former and the latter, respectively. The K − S test shows significantly different distributions ofX-ray photon index in two regions with D = 0 .
440 and P = 0 . −
10 keV is shownin Figure 12. Indeed, the PSF effect results in flatter spectra in outer region than inner region, which even is moresignificant than the observational result. Therefore, by taking this PSF effect into account, the X-ray spectra at outerregion is slightly steeper than inner region. However, the firm conclusion can’t be drawn due to the large uncertaintiesin the photon index. REFERENCES
Aars, C. E., Hough, D. H., Yu, L. H., et al. 2005, AJ, 130,23Abazajian, K. N., Adelman-McCarthy, J. K., Ag¨ueros,M. A., et al. 2009, ApJS, 182, 543Alam, S., Albareti, F. D., Allende Prieto, C., et al. 2015,ApJS, 219, 12Alexander, D. M., Stern, D., Del Moro, A., et al. 2013,ApJ, 773, 125Aldcroft, T. L., Bechtold, J., & Elvis, M. 1994, ApJS, 93, 1Aldcroft, T. L., Elvis, M., & Bechtold, J. 1993, AJ, 105,2054Aldcroft, T. L., Siemiginowska, A., Elvis, M., et al. 2003,ApJ, 597, 751 Anderson, S. F., Voges, W., Margon, B., et al. 2003, AJ,126, 2209Antonucci, R. 1993, ARA&A, 31, 473Ballantyne, D. R., Ross, R. R., & Fabian, A. C. 2002,MNRAS, 332, L45Bedford, N. H., Kerr, A. J., Mathur, S. H., et al. 1981,MNRAS, 195, 245Belsole, E., Worrall, D. M., & Hardcastle, M. J. 2006,MNRAS, 366, 339Blandford, R., Meier, D., & Readhead, A. 2019, ARA&A,57, 467Boissay, R., Ricci, C., & Paltani, S. 2016, A&A, 588, A70Boksenberg, A., Carswell, R. F., & Oke, J. B. 1976, ApJL,206, L121 Zhou & Gu
Table 4.
The X-ray spectral fitting for different regions . (cid:48)(cid:48) circle 0 . − . (cid:48)(cid:48) annulusName CXO ID Γ Γ χ Red /dof.
Stat. Γ χ Red /dof.
Stat.(1) (2) (3) (4) (5) (6) (7) (8) (9)3C 9 1595 1 . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . . +0 . − . Note —Column (1): 3CRR name; Column (2): Chandra observation ID; Column (3): Photon indexof power-law component extracted from source-centered 2 . (cid:48)(cid:48) circle; Column (4 − χ /degree of freedom and statistical method for the power-law fit in 1 . (cid:48)(cid:48) circle; Columns(7 − . − . (cid:48)(cid:48) annulus. -ray emissions of 3CRR quasars N Figure 11.
The distribution of X-ray photon index Γ. Thesolid line is for 2 . (cid:48)(cid:48) region, and the dashed line represents1 . (cid:48)(cid:48) region. As comparison, the 0 . − . (cid:48)(cid:48) annulus is shownas dot-dashed line. ( . . " ) Simulation1.0 1.5 2.0 2.5 (1.5")0123 ( . . " ) Observation
Figure 12.
The relation of the X-ray photon index from1 . (cid:48)(cid:48) circle and from 0 . − . (cid:48)(cid:48) annulus: ( upper ) - simulation;( bottom ) - observation. The black dotted line is equivalentline, and its 0.5 offset is shown as dashed lines. The blacksolid lines show the linear fit.Brandt, W. N., & Alexander, D. M. 2015, A&A Rv, 23, 1Brandt, W. N., & Hasinger, G. 2005, ARA&A, 43, 827Bregman, J. N., Glassgold, A. E., Huggins, P. J., & Kinney,A. L. 1985, ApJ, 291, 505Bridle, A. H., Hough, D. H., Lonsdale, C. J., Burns, J. O.,& Laing, R. A. 1994, AJ, 108, 766Brotherton, M. S., Wills, B. J., Dey, A., van Breugel, W., &Antonucci, R. 1998, ApJ, 501, 110 Brotherton, M. S., Ly, C., Wills, B. J., et al. 2002, AJ, 124,1943Brunetti, G., Bondi, M., Comastri, A., et al. 2001, ApJL,561, L157Brunetti, G., Bondi, M., Comastri, A., & Setti, G. 2002,A&A, 381, 795Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989,Interstellar Dust, 135, 5Cohen, M., Wheaton, W. A., & Megeath, S. T. 2003, AJ,126, 1090Crawford, C. S., & Fabian, A. C. 2003, MNRAS, 339, 1163Croston, J. H., Hardcastle, M. J., Harris, D. E., et al. 2005,ApJ, 626, 733Crummy, J., Fabian, A. C., Gallo, L., & Ross, R. R. 2006,MNRAS, 365, 1067Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003,The IRSA 2MASS All-Sky Point Source CatalogDavis, J. E. 2001, ApJ, 562, 575Dewangan, G. C., Griffiths, R. E., Dasgupta, S., & Rao,A. R. 2007, ApJ, 671, 1284Dickey, J. M., & Lockman, F. J. 1990, ARA&A, 28, 215Donahue, M., Daly, R. A., & Horner, D. J. 2003, ApJ, 584,643Donato, D., Ghisellini, G., Tagliaferri, G., & Fossati, G.2001, A&A, 375, 739Done, C., Davis, S. W., Jin, C., Blaes, O., & Ward, M.2012, MNRAS, 420, 1848Elvis, M., Wilkes, B. J., McDowell, J. C., et al. 1994, ApJS,95, 1Erlund, M. C., Fabian, A. C., Blundell, K. M., Celotti, A.,& Crawford, C. S. 2006, MNRAS, 371, 29Fabian, A. C., Ballantyne, D. R., Merloni, A., et al. 2002,MNRAS, 331, L35Fabian, A. C., Celotti, A., & Johnstone, R. M. 2003,MNRAS, 338, L7Fan, J. H., & Zhang, J. S. 2003, A&A, 407, 899Fanaroff, B. L., & Riley, J. M. 1974, MNRAS, 167, 31PFernini, I., Burns, J. O., & Perley, R. A. 1997, AJ, 114, 2292Fernini, I. 2007, AJ, 134, 158Fernini, I. 2014, ApJS, 212, 19Fiore, F., Elvis, M., Mathur, S., Wilkes, B. J., & McDowell,J. C. 1993, ApJ, 415, 129Freeman, P., Doe, S., & Siemiginowska, A. 2001,Proc. SPIE, 4477, 76Fruscione, A., McDowell, J. C., Allen, G. E., et al. 2006,Proc. SPIE, 6270, 62701VGambill, J. K., Sambruna, R. M., Chartas, G., et al. 2003,A&A, 401, 505Garc´ıa, J., Dauser, T., Lohfink, A., et al. 2014, ApJ, 782, 76 Zhou & Gu
Gaskell, C. M., Goosmann, R. W., Antonucci, R. R. J., &Whysong, D. H. 2004, ApJ, 616, 147Goulding, A. D., Zakamska, N. L., Alexandroff, R. M., etal. 2018, ApJ, 856, 4Grandi, P., Malaguti, G., & Fiocchi, M. 2006, ApJ, 642, 113Grandi, P., & Palumbo, G. G. C. 2004, Science, 306, 998Green, P. J., Schartel, N., Anderson, S. F., et al. 1995, ApJ,450, 51Grimes, J. A., Rawlings, S., & Willott, C. J. 2005,MNRAS, 359, 1345Gu, M., Cao, X., & Jiang, D. R. 2001, MNRAS, 327, 1111Gupta, M., Sikora, M., Rusinek, K., & Madejski, G. M.2018, MNRAS, 480, 2861Gupta, M., Sikora, M., & Rusinek, K. 2020, MNRAS, 492,315G¨urkan, G., Hardcastle, M. J., & Jarvis, M. J. 2014,MNRAS, 438, 1149Haardt, F., & Maraschi, L. 1993, ApJ, 413, 507Haas, M., M¨uller, S. A. H., Bertoldi, F., et al. 2004, A&A,424, 531Haas, M., Willner, S. P., Heymann, F., et al. 2008, ApJ,688, 122-127Hardcastle, M. J., Birkinshaw, M., Cameron, R. A., et al.2002, ApJ, 581, 948Hardcastle, M. J., Evans, D. A., & Croston, J. H. 2006,MNRAS, 370, 1893Hardcastle, M. J., Evans, D. A., & Croston, J. H. 2009,MNRAS, 396, 1929Hardcastle, M. J., Harris, D. E., Worrall, D. M., &Birkinshaw, M. 2004, ApJ, 612, 729Harris, D. E., & Krawczynski, H. 2006, ARA&A, 44, 463Harris, D. E., Massaro, F., & Cheung, C. C. 2010, X-rayAstronomy 2009; Present Status, Multi-wavelengthApproach and Future Perspectives, 355Healey, S. E., Romani, R. W., Taylor, G. B., et al. 2007,ApJS, 171, 61Jin, C., Done, C., & Ward, M. 2017, MNRAS, 468, 3663Kalberla, P. M. W., Burton, W. B., Hartmann, D., et al.2005, A&A, 440, 775Kataoka, J., Stawarz, (cid:32)L., Takahashi, Y., et al. 2011, ApJ,740, 29Kellermann, K. I., Sramek, R., Schmidt, M., et al. 1989,AJ, 98, 1195Krawczyk, C. M., Richards, G. T., Mehta, S. S., et al. 2013,ApJS, 206, 4Laing, R. A., Riley, J. M., & Longair, M. S. 1983, MNRAS,204, 151Laurent-Muehleisen, S. A., Kollgaard, R. I., Ryan, P. J., etal. 1997, A&AS, 122, 235 Lehnert, M. D., Miley, G. K., Sparks, W. B., et al. 1999,ApJS, 123, 351Li, S.-L. 2019, MNRAS, 490, 3793Lusso, E., Comastri, A., Vignali, C., et al. 2010, A&A, 512,A34Lusso, E., Hennawi, J. F., Comastri, A., et al. 2013, ApJ,777, 86Madsen, K. K., F¨urst, F., Walton, D. J., et al. 2015, ApJ,812, 14Magdziarz, P., Blaes, O. M., Zdziarski, A. A., Johnson,W. N., & Smith, D. A. 1998, MNRAS, 301, 179Mallick, L., & Dewangan, G. C. 2018, ApJ, 863, 178Mallick, L., Dewangan, G. C., Gandhi, P., Misra, R., &Kembhavi, A. K. 2016, MNRAS, 460, 1705Marshall, H. L., Edelson, R. A., Vaughan, S., et al. 2003,AJ, 125, 459Marshall, H. L., Schwartz, D. A., Lovell, J. E. J., et al.2005, ApJS, 156, 13Maselli, A., Kraft, R. P., Massaro, F., et al. 2018, A&A,619, A75Massaro, F., Cheung, C. C., & Harris, D. E. 2010, X-rayAstronomy 2009; Present Status, Multi-wavelengthApproach and Future Perspectives, 475Massaro, F., Cheung, C. C., & Harris, D. E. 2011a, Jets atAll Scales, 160Massaro, E., Giommi, P., Leto, C., et al. 2009, A&A, 495,691Massaro, F., Harris, D. E., & Cheung, C. C. 2011b, ApJS,197, 24Massaro, F., Harris, D. E., Chiaberge, M., et al. 2009, ApJ,696, 980Massaro, E., Maselli, A., Leto, C., et al. 2015, Ap&SS, 357,75Massaro, F., Missaglia, V., Stuardi, C., et al. 2018, ApJS,234, 7Massaro, E., Perri, M., Giommi, P., et al. 2004, A&A, 413,489Massaro, F., Harris, D. E., Tremblay, G. R., et al. 2010,ApJ, 714, 589Massaro, F., Harris, D. E., Tremblay, G. R., et al. 2013,ApJS, 206, 7Massaro, F., Harris, D. E., Liuzzo, E., et al. 2015, ApJS,220, 5Massaro, F., Tremblay, G. R., Harris, D. E., et al. 2012,ApJS, 203, 31Mathur, S., Wilkes, B., Elvis, M., & Fiore, F. 1994, ApJ,434, 493McLure, R. J., Jarvis, M. J., Targett, T. A., Dunlop, J. S.,& Best, P. N. 2006, MNRAS, 368, 1395 -ray emissions of 3CRR quasars23