The Jet of 3C 17 and the Use of Jet Curvature as a Diagnostic of the X-ray Emission Process

0.33 MB·English

by  F. Massaro

version 2009Jan26: fm/deh
PreprinttypesetusingLATEXstyleemulateapjv.10/09/06
THE JET OF 3C 17 AND THE USE OF JET CURVATURE AS A DIAGNOSTIC OF THE X-RAY EMISSION
PROCESS
F. Massaro1, D. E. Harris1, M. Chiaberge2,7, P. Grandi3, F. D. Macchetto2, S. A. Baum4, C. P. O’Dea5 and
A. Capetti6
version 2009 Jan 26: fm/deh
ABSTRACT
9 WereportontheX-rayemissionfromtheradiojetof3C17fromChandraobservationsandcompare
0
the X-ray emission with radio maps from the VLA archive and with the optical-IR archival images
0
from the Hubble Space Telescope. X-ray detections of two knots in the 3C 17 jet are found and both
2
of these features have optical counterparts. We derive the spectral energy distribution for the knots

n in the jet and give source parameters required for the various X-ray emission models, finding that
a both IC/CMB and synchrotron are viable to explain the high energy emission. A curious optical
J
feature (with no radio or X-raycounterparts)possibly associatedwith the 3C 17 jet is described. We

9 also discuss the use of curved jets for the problem of identifying inverse Compton X-ray emission via
2 scattering on CMB photons.

Subject headings: Galaxies: active — galaxies: jets — galaxies: individual (3C 17)— X-rays: general
]
E — radio continuum: galaxies — radiation mechanisms: nonthermal
H
h. 1. INTRODUCTION 1997). This source shows also a significant optical po-
p The X-ray radiation observedfrom radio jets is gener- larization in its nucleus (Tadhunter et al. 1998), and its
- first detection in X-rays has been reported by Siebert et
ally interpreted to be from non-thermal processes, even
o
al. (1996) using the ROSAT All Sky survey data.
r if its nature is still unclear for any particular jet. It
t could be described in terms of synchrotron emission or Here, we report the major results concerning the mul-
s tiwavelength studies of the jet in 3C 17. We present the
a in terms of several varieties of inverse Compton radia-
X-ray data of this source together with the optical-IR
[ tion. So to understand the emission mechanisms related
images (HST) and the radio maps (VLA archive).
tothesecomponentsthemultiwavelengthapproachisre-
1 For our numerical results, we use cgs units unless
quired. If the X-ray emission is synchrotron, electrons
v with Lorentz factors γ up to 107 are required whereas stated otherwise and we assume a flat cosmology with
8 if the process is inverse Compton radiation with seed H0 = 72 km s−1 Mpc−1, ΩM = 0.27 and ΩΛ = 0.73
1 ′′
(Spergel et al., 2007), so 1 is equivalent to 3.47 kpc.
47 pXh-roatyosnswoduuledtcoomtheefrCoMmBele(cTtarvoencscwhiiothetγal.,120000(0H),artrhies Spectralindices,α,aredefinedbyfluxdensity,Sν ∝ν−α.
. & Krawczynski 2002). To investigate the≈nature of the 2. OBSERVATIONSANDDATAREDUCTION
1
0 emission in jets we analyze the jet of the powerful radio 2.1. X-rays data
galaxy 3C 17.
9
3C 17 was observed during the first year of the Chan- 3C 17 has been observed by Chandra (Obs ID 9292)
0
dra 3C snapshot program, which started in AO-9 with on February 2, 2008, with the ACIS-S camera, oper-
:
v 8ks observations of 30 of the previously unobserved (by ating in VFAINT mode, with an exposure of about 8
Xi Chandra) 3C sources with z<0.3. The 3C sample allows ksecs. The data reduction has been performed following
us to have multifrequency data available from the HST the standard procedures described in the Chandra In-
ar andthe VLAarchives. 3C 17 is a radiogalaxy(z 0.22, teractive Analysis of Observations (CIAO) threads and
Schmidt et al. 1965) with a peculiar radio structu∼re in- using the CIAO software package v3.4. The Chandra
vestigated by Morganti et al. (1999). Its Hα emission Calibration Database (CALDB) version 3.4.2 was used
hasastrongbroadcomponentandboththe[OII]λ3727 to process all files. Level 2 event files were generated
and [O III] λ5007 emission lines are extended (Dickson using the acis process events task, after removing the
hot pixels with acis run hotpix. Events were filtered
1Harvard, Smithsonian Astrophysical Observatory, 60 Garden forgrades0,2,3,4,6andweremovedpixelrandomization.
Street,Cambridge,MA02138 Astrometricregistrationwasdonechangingtheappropi-
2Space Telescope Science Institute, 3700 San Martine Drive, ate keywordsin the fits headerso as to alignthe nuclear
Baltimore,MD21218
4CarlsonCenter forImaging Science 76-3144, 84 LombMemo- X-raypositionwiththatofthe radio. We alsoregistered
rialDr.,Rochester, NY14623 the HST images in the same way.
3INAF-IASF-IstitutodiAstrofisicaSpazialeefisicacosmicadi We created 3 different fluxmaps (soft, medium, hard,
Bo5loDgnepat,VoifaPPh.yGsoicbse,ttRio1c0h1e,s4t0er129In,sBtiotluotgenao,fItTaelychnology, Carl- in the ranges 0.5 – 1, 1 – 2, 2 – 7 keV, respectively)
by dividing the data with the exposure maps. When
son Center for Imaging Science 76-3144, 84 Lomb Memorial Dr.,
Rochester,NY14623 constructing the fluxmaps, we normalized eachcount by
6INAF - Osservatorio Astronomico di Torino, Strada Osserva- multiplying by hν where ν corresponds to the energy
torio20,I-10025PinoTorinese,Italy
7INAF - Istituto di Radioastronomia di Bologna, via Gobetti used for the corresponding exposure map. Thus we can
measurethefluxinanyapertureincgsunitswithonlya
10140129Bologna,Italy
small correction for the ratio of the mean energy of the
2 Massaro,F., Harris, D. E., Chiaberge, M., Grandi, P., Macchetto, F. D., et al.
countswithinthe aperturetothe nominalenergyforthe
TABLE 1
band.
ObservedX-ray countsandfluxesforjet features.
Photometric apertureswere constructed so as to acco-
modate the Chandra point spread function and so as to
Soft Medium Hard Total
include the total extent of the radio structure. They are
shown in fig. 1. The background regions have been cho- NominalEnergy(keV) 0.70 1.4 4.0
sen close to the source with comparable size, typically Band(keV) 0.5-1.0 1-2 2-7 0.2-7
two times bigger than the source region, centered on a S3.7counts 5 6 1 12
position where other sources or extended structure are S11.3counts 1 0 4 5
notpresent. AllX-rayflux densitieshavebeencorrected S3.7flux 2.6±1.1 2.2±0.9 1.6±1.6 6.4
fortheGalacticabsorptionwiththeNH columndensities S11.3flux 0.6±0.5 0 6.5±3.3 7.1
given by Kalberla et al. (2005), 2.861020cm−2. Fluxunits: 10−15 ergcm−2 s−1.
· Notestotable:
2.2. Radio maps and HST images Theaveragebackgrounds measuredforthetotal 0.5-7keVband
fromannularringsaroundtheradiogalaxyare1.57counts (S3.7)
We compare our X-rays maps of 3C 17 with the VLA and0.08counts (S11.3).
radio data described in Morganti et al. (1999) at 4.8
′′
GHz with a beamsize of 0.4 . We also reduced archival Thenucleusofthehostgalaxyhasbeenobservedwith
VLA data at 1.54 GHz and 14.9 GHz with AIPS stan- the Very Long Baseline Array by Venturi et al. (2000)
dard reduction procedures. The angular resolution of whodescribe3C17asa“transitionobject”betweenFRI
′′
theseradiomapsis 1.4 andthefinalimageisingood and FRII. The pc scale jet shows a ’core’ with an exten-
agreement with the≈4.8 GHz radio map. The amplitude sion in PA 100◦ to 110◦, followed by lower brightness
≈
calibratorused at1.4 GHz was3C 48 andthe phase cal- features. This position angle is essentially the same as
ibrator was 0056-001. At 15GHz we used 0106+103 for that of the first kpc scale knot, S3.7, discussed below.
both amplitude and phase. Thekpcscaleradiomorphologyisdominatedbyasin-
Concerning the other bands, we compared our X- glesided,stronglycurvedjet(fig.1,asdescribedbyMor-
ray image to the IR HST observation at 1.6 µm ganti et al. 1993, 1999), although there is lower bright-
(1.871014 Hz, H band)8, and to the STIS visible im- nessemissionoutsidetheareacoveredinthefigure. Sim-
·
age at 4.161014 Hz (7216 ˚A, R band). We also used a ilarly to M87, 3C 17 was originally classified as a ’core-
FUV HST i·mage at 1457˚A. halo’ source. The jet has a bright knot at 3.7′′ from the
′′
nucleus while the curved part lies at about 11 from the
nucleus.
32.0
3.1. Jet knots
34.0
36.0 41.0
38.0
42.0
-2:07:40.0
NUCLEUS
S3.7
42.0
43.0
44.0
46.0 -2:07:44.0
48.0 S11.3
45.0
50.0
21.4 21.2 21.0 20.8 20.6 20.4 20.2 0:38:20.0 19.8
46.0
0.2 0.6 1
Fig. 1.—The5GHzVLAmapofMorgantietal. (1999)witha 47.0
restoringbeam of 0.4′′. The twoknots of interest areemphasized
(inwhite)withtheregionsusedforphotometry. 21.3 21.2 21.1 0:38:21.0 20.9 20.8 20.7
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
3. RESULTS reFmigo.ve2d.—anTdhtheeCehvaenndtsrabXet-wraeyenm0a.5p.anPdix7elkreaVndwoemreizsamtiooonthheadsbweietnh
3C 17 is a broad lined radio galaxy (BLRG) aGaussianofFWHM=0.87′′. Theradiocontourscomefromfig.1
and start at 0.5 mJy/beam, increasing by factors of 4. S3.7 is at
(Buttiglione et al. 2008) with a monochromatic radio
theupperrightandS11.3tothelowerleft.
luminosity, log P1.4GHz 26.9 which is at least two or-
≈
ders of magnitude above the division between FRI and
UsingournewChandraobservationwefinddetections
FRII types(Ledlow&Owen,1996)butwithanambigu-
of two knots (fig. 2). The first knot, S3.7, lies at a pro-
ous radio morphology. Miller & Brandt (2008) provide
jected distance of 12.8 kpc from the nucleus and is re-
a more extensive discussionon classifyingsourcesof this
solved with the VLA. The deconvolved size (FWHM) is
’hybrid’type,including4C65.15,whichisverysimilarto
0.46′′ 0.18′′ (1.6 0.6 kpc) in PA=115◦. Following a
3C 17 in many respects. × ×
gap with no detectable radio emission, the jet again be-
′′
8 availableon theHST Snapshots of 3CR Radio Galaxies web- comesvisibleinthe radioabout7 fromthe nucleusand
page,http://archive.stsci.edu/prepds/3cr/ brightensasitapproachestheregionofmaximumappar-
The Jet of 3C 17 and the Use of Jet Curvature as a Diagnostic of the X-ray Emission Process 3
We tried several models to fit the emitted spectrum
TABLE 2
of both knots in 3C 17. We describe the spectrum from
Flux densities (cgs units) for3C17 knots
the radio band to the UV in terms of synchrotron emis-
sionbutconsiderbothsynchrotronandIC/CMBfor the
X-rays. Weperformedourcalculationsassumingthefol-
Freq. (Hz) Band S3.7 S11.3
lowing hypothesis: (1) the distribution of emitting elec-
1.66·109 L 80±10·10−26 (190±10·10−26) trons is a power-law with slope s; (2) the volumes of
4.86·109 C 30±1·10−26 83±2·10−26 the emitting regions correspond to the deconvolved ra-
1.49·1010 U 12.4±1·10−26 (33±10·10−26)
dio sizes; (3) the magnetic field is in equipartition with
1.87·1014 1.6µm 3.28±0.16·10−29 ...
4.16·1014 7216˚A 1.44±0.13·10−29 1.03±0.07·10−29 the energy density of the relativistic electrons; and (4)
2.06·1015 1457˚A 0.15±0.03·10−29 0.054±0.018·10−29 the proton-to-electronratio is assumed to be zero.
1.18·1017 softX 1.86±0.8·10−32 0.56±0.27·10−32 Basedontheseassumptions,thespectrumisdescribed
3.39·1017 mediumX 1.04±0.42·10−32 <0.3·10−32 in terms of 4 parameters, namely: the slope of the elec-
9.67·1017 hardX 0.14±0.14·10−32 0.56±0.28·10−32 tron distribution s, the maximum and the minimum en-
Notestotable ergy of emitting particles γmax,γmin and the magnetic
Valuesinparentheses areuncertainbecausea1.4′′ beamsizeis field,becausewefixedthevolumederivedfromtheradio
inadequate toisolateS11.3fromadjacentknots. maps.
Thespectralindexoftheelectrondistributionhasbeen
ent curvature. It is at this point we detect X-rays from derived from the observed spectrum fitting the radio to
theradioknot,S11.3,whichhasthehighestradiosurface optical data with a power-law (see Tab.3). The maxi-
brightness (after the nuclear emission). Its deconvolved mum energy of particles has been evaluated in order to
radio size is 0.4′′ 0.3′′ (1.4 1.0 kpc) in PA=48◦. seethesynchrotronexponentialcut-offintheUV,assug-
× ×
X-rayfluxesforbothknotsaregiveninTable1andflux gested by the data. Finally, given the value of the mag-
densities, evaluated for a spectral index equal to 1, are netic field and assuming a minimum observed frequency
listed in Table 2. The total number of counts detected of 107Hz, we derived the γmin parameter for both elec-
for S3.7 and for S11.3 is 12 and 5 respectively, where trondistribution (see Tab. 3). Following this criteriawe
the average background evaluated around each knot is found that the γmin is the order of 100 for both knots,
1.57 cts for S3.7 and 0.08 cts for S11.3 for the same size it correspond to an electron minimum energy of about
aperture. 50 MeV, that can be predicted by several acceleration
processes (e.g. Protheroe 2004). Finally, the values of
-2:07:40.0 the magnetic field, reported in Tab.3, are related to the
S3.7 synchrotroninterpretation,butareonlyslightlydifferent
41.0
for the smaller γmin.
Linear Object
42.0 Solutionsshowninfigs.4 &5havethe beaming factor
fixedto1. Wefindagoodagreementofoursolutionwith
43.0
theobservedspectrumforamagneticfieldof180µGfor
44.0 S3.7 and 195 µG for S11.3. Parameters for our model
are reported in Table 3.
45.0 For an inverse Compton (IC) model using photons of
the microwavebackground(CMB) we follow the formal-
46.0
ism of Harris & Krawczynski (2002). For S3.7, we find
47.0 that the required beaming factor is δ=8 for the fiducial
condition δ = Γ (Γ is the bulk Lorentz factor of the jet
48.0 S11.3
knot). The angle to the line of sight for this solution
49.0 is θ=7◦ although smaller angles together with smaller Γ
21.3 21.2 21.10:38:21.0 20.9 20.8 20.7 20.6 20.5 20.4 arealsoacceptable. ForS11.3,the values forthe fiducial
Fig. 3.— HST optical image of 3C 17 (7216˚A). The overlaid conditionareδ =Γ=5.5andθ=10◦. Sincethejetisobvi-
cyancontoursarethesameasthoseinfig.2. Here,theradioknots ouslycurved,thereisno expectationthatthe twovalues
at 3.7′′ and 11.3′′ show optical counterparts. The linear object of θ should be equal even if one might guess that S11.3
discussedinthetextliesatabout7.4′′ fromthenucleus.
wouldbemovingmoretowardsusthanS3.7. Theseesti-
mates are very roughbecause of the poor signalto noise
We compared also the radio and the X-ray emission
inthe X-raydataandbecausewedonothaveameasure
with the optical and IR images reported by Donzelli et
of the radio spectral index.
al. (2007). The knot S3.7 shows a counterpart both in
It should be noted that the IC/CMB interpretation
the optical and in the IR image. For the other knot,
rests solely on the UV flux densities which are approxi-
S11.3,wereportonlythe opticalassociationbecausethe
mately 5 sigma for S3.7 and less than 3 sigma for S11.3
FOV of the NICMOS camera is too small. The IR to
below the single power-lawextending to the X-ray data.
UV flux densities are corrected for the reddening using
Because of the low statistical significance of these data
the following values of absorption: AH=0.01, AR=0.062
and because of the added uncertainty of the extinction
and AUV=0.044, for the IR, the R band and the UV
correction, we do not rule out a synchrotron spectrum
respectively (Cardelli et al. 1989)9. The SED of each
extendingouttothe X-raysasshownbythedottedlines
knot is shown in figs. 4 & 5.
in figures 4 & 5. However the low UV flux values fa-
9 see also “Doug’s Excellent Absorption Law Calculator”, vortheIC/CMBinterpretationinsteadofthesinglesyn-
http://wwwmacho.mcmaster.ca/JAVA/Acurve.html chrotroncomponent. Amoreaccurateobservationinthe
4 Massaro,F., Harris, D. E., Chiaberge, M., Grandi, P., Macchetto, F. D., et al.
UV band (down to 1200˚A) is needed to distinguish
∼ TABLE 3
between models, e.g. if the spectrum is curved in the
Model parametersforthesynchrotroncalculations.
IR-to-UV range then the single synchrotron component
could be ruled out.
Parameter(cgs units) S3.7 S11.3
Spectral Index,αro 0.87 0.99
Electronindexs=2α+1 2.74 2.98
γmin 104 100
γmax(IC/CMB) 1.8·106 2.2·106
γmax(synchrotron) 3.1·107 2.8·107
10-24 Radio Magneticfield(10−6 G) 180 195
Volume(1064cm3) 1.81 4.21
Luminosity(1043 ergs−1) 1.36 0.98
10-26
Notestotable
-1Hz) a)Twovaluesofγmax arelisted. TheIC/CMBentryisforthe
-2-1 (erg cm s 1100--3208 IR opticaUlV cwebsyax)hnsteTeecrnhhewdraehoisnvtergtaroheltuneotehsisntyehtonesefcyrhtXpnhrrc-oeehrttarmrayootatnibrogoannennn.edttsri.pcyefiicsetlrfduomrreapcsuoyrtntsecodhffraoratefrtorenerlatshpteeedcUttrVoumtdhaeta
Fν
X-rays
10-32
TABLE 4
Flux densities forthelinearobject.
10-34
108 109 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019
ν (Hz) Frequency Fluxdensity
5GHz <0.5mJy
1.6µm 4.8±0.3µJy
Fig. 4.— The spectrum of S3.7. The synchrotron calcula- 7216˚A 0.73±0.04µJy
tions (done under the assumptions described in §3) fit the radio
1457˚A <0.15nJy
to UV with a cutoff, or extend to the X-ray ignoring the UV da-
tum(dashedline). TheIC/CMBmodelistheseparatecomponent 1×1018Hz <0.1nJy
peakingjustbelow1017Hz.
just the point where the radioemission recommencesaf-
ter the ’gap’ following S3.7. This feature is 7.3′′ (a
≈
projected distance of 25kpc) from the nucleus and has
an optical AB magnitude of 23. At 7200 ˚A, there is a
≈
’hole’inthecenter;oneormorepixelsarecomparableto
10-24 Radio the background level. At 1.6µm, there is no hole. If we
measure just the outer bits, we find a two-pointspectral
index,αo=1.7 0.2. Ifwemeasuretheentireobject,itis
±
10-26 2.2 0.2. There is no evidence ofX-rayor radioemission
-1Hz) cor±responding to this feature. Flux densities and upper
-2-1m s 10-28 IR optical limIfittshaereobgjievcetnwinerTeaabnlee4d.ge on spiral at the same dis-
(erg c10-30 UV twahnicceh,ofwh3Cen1c7o,uiptlsedabwsoitlhuteanmoavgenriatulldesizweouofld3b.5e k-p17c
Fν would mean it could be classified as a dwarf spiral (see
X-rays Schombert et al. 1995).
10-32 We consider 4 types of possibile explanations for this
object.
10-34108 109 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 • The object is a foreground or background object
ν (Hz) (e.g. edge on spiral) and has nothing whatso-
ever to do with the jet. A rough estimate of
the probability of the jet crossing a random back-
Fig. 5.—ThespectrumofS11.3Seethecaptionoffig.4. ground source within the 0.2′′ nuclear region (de-
fined bythe loweredbrightnesscenterat7216˚A)is
0.2/360=5 10−4. However, since we don’t know
×
how to estimate the probability that the (up-
3.2. A peculiar optical feature
stream) invisible jet just happens to start convert-
We noticed a linear optical feature on both HST im- ingsomeofitspowerintorelativisticelectronsand
ages: 1.6 µm and 7216 ˚A (fig. 3). It is a little over an B field at this location, and since we are wary of a
arcsec long (3.5kpc if at the distance of 3C 17), with a posteriori probabilities, the ’chance alignment’ hy-
◦
major axis within 11 of the perpendicular to the jet at pothesis seems unlikely, but remains a possibility.
The Jet of 3C 17 and the Use of Jet Curvature as a Diagnostic of the X-ray Emission Process 5
The emitting region arises from the interaction of the ratio may decrease smoothly (mimicking the syn-
•
the jet and some pre-existing entity (e.g. a large chrotron) or is sensibly constant down the jet (“class 2”
HI cloud, only a part of which gets ionized and e.g. 4C19.44,Schwartz et al. 2007). Normally these two
produces free-free emission). This hypothesis can possibilities are ascribedto a smoothly decelerating flow
be tested with an optical spectrum since the most (thereby diminishing the IC component) or a relatively
likelymethodofcreatingopticalemissionfromthe uniform value for Γ which would maintain the effective
interactionofajetandcoldgaswouldbeviaioniza- energy density of the CMB.
tionleadingtorecombinationlinesandanoptically Foracurvedjetwithchangingθ,theremaybeanoma-
thincontinuum. Sincetheobservedspectrumisin- lous changes in synchrotron brightness associated with
consistent with free-free emission from an ionized the effect of θ on δ (the beaming factor), but the ra-
gas at > 104K, it would have to be dominated by tio of X-ray to radio emission should be preserved and
emissionlines. Thetwoobservedbandscorrespond remain unaffected by changing θ. In the IC/CMB sce-
to1.27-1.36µand5452-6387angstromsatthered- nario, the critical point is that a smaller θ will lead to a
shift of 3C 17. Neither of these bands would be markedchange in R (unlike the synchrotroncase), devi-
expected to contain the more likely emission lines ating either from the smoothly decreasing ratio or from
envisaged by this scenario. a constant ratio.
The knotS11.3is brighterthanadjacentknotsin all3
Theemittingregioncomesfromanunknownprop- bands (radio, optical, and X-rays). Is this because it is
•
erty of the jet. To our knowledge, no other jet wherethe curvingjet lies closestto the l.o.s. or is itjust
exhibits such a narrow band feature perpendicular an intrinsically stronger emitting region which might be
to its axis. causedbyalongerpathlengthatatangentialpoint? The
longer pathlength possibility will again not disturb the
Theobjectisindeedanedgeonspiralandisaclose intrinsic X-ray/radio behavior for either emission mech-
•
companion of 3C 17. The jet pierces the center of anism, so if we can show that S11.3 has an anomalously
this galaxy and that is why the jet begins to be large ratio, it will be a strong indicator that the X-ray
visible at this location. This jet, like all one sided emission process is IC/CMB.
◦
jets, is coming ’mostly’ towards us: perhaps 10 - For IC/CMB emission, the preference for IC scatter-
◦
30 to the l.o.s. for this section of the jet. Since ing when the electrons are meeting the photons ’head-
the edge on spiral’s plane is perpendicular to the on’ in the jet frame, produces more IC emission in the
plane of the sky, the actual impacting jet will be downstream direction. The angular dependency of this
closetohittingtheplaneofthegalaxyatanoblique extra beaming term is given in eq. (A22) of Harris &
angle, not coming in at the pole, as it appears in Krawczynski(2002) and with a few substitutions can be
the projectedview. Inanyeventthe probabilityof written as:
hitting an object whose’s center subtends 1 kpc2
(as seen from the SMBH of 3C 17) by chance is
4π1R2 ≤ 1.3×10−4 (again, an a posteriori proba- µΓ √Γ2 1 2
bility). ξ = 1+ − − (1)
( Γ µ√Γ2 1)
− −
If we refuse to allow ’intent’ (e.g. ’intergalactic engi-
neering’), we are left with an improbable chance align- where Γ is the bulk Lorentz factor of the emitting re-
ment, an interaction with some pre-existing entity, or gion and µ=cosθ. This function is shown in fig. 6 for a
some new type of jet-related emission. An optical spec- few representative values of Γ.
trum of this object could eliminate some of these possi- Although our data for 3C 17 are inadequate to per-
bilities. form a meaningful test, current parameters are given in
Table 5 as an illustration of the method. While it is
4. HOWCURVEDJETSCANPROVIDEEVIDENCEFOR true that R is larger for S11.3 than for the adjoining
IC/CMBX-RAYEMISSION knots, R(S3.7) is more than twice R(S11.3). To sustain
Most jets that have been well studied are relatively an IC/CMB explanation for the X-raysof both knots, it
straight and the assumption is normally made that θ, becomes necessary to posit a smaller value of Γ for the
the angle to the l.o.s., does not change along the jet. A outer parts of the jet comparedto that ascribed to S3.7.
curving jet provides an advantage because a changing That would mean that the Γ2 term in eq.(A22) (ibid)
viewing angle should be reflected in the run of the ratio, woulddominatethechangeinRbetweenS3.7andS11.3.
R,ofX-raytoradiointensitiesdifferentlyforsynchrotron In the present context the beaming parameters for S3.7
and IC emissions. If, and only if, the X-ray emission is could be (see 3) Γ=δ=8, θ=7◦ while at S11.3,Γ=5.4,
dominated by IC/CMB emission will the emitting re- δ=5.4, and θ=§10◦ and in this case, the angular depen-
gions closer to the line of sight than the others display dence is not the dominant effect. Obviously we have too
anomalously large values of R. muchfreedombecauseoftheshortChandraobservation:
In X-ray synchrotron jets, R is often a sharply de- whatisrequiredforthistestisalongerobservationwhich
creasing function of distance down the jet (“class 1” wouldproviderobustX-raydetectionsofalltheknotsin
e.g. 3C273, Jester et al. 2006), and we normally as- Table 5 so as to compare R values all along the jet.
cribe this to a decreasing ability to produce electrons of Othersourceswithcurvedjets are3C120(forwhichit
the required energies (possibly caused by an increasing is not obvious where θ is at a minimum) and 4C65.15,
magnetic field strength). In other straight jets normally a higher redshift quasar whose morphology mimics that
thoughttoberepresentativeofIC/CMBX-rayemission, of 3C 17 and also has an X-ray detection of a knot at
6 Massaro,F., Harris, D. E., Chiaberge, M., Grandi, P., Macchetto, F. D., et al.
5. SUMMARY
Wehavedetectedtwoknotsinthe3C17jetinbothX-
raysand optical/IRbands. The resulting radio to X-ray
Γ = 3 spectra do not provide a definitive answer to the choice
between IC/CMB and synchrotron X-ray emission. Ad-
1
ditionally, we have described a peculiar optical object,
o Γ = 6 possibly an edge-on spiral galaxy, which appears to be
nc rati Γ = 10 pfiiceiercnetdfobrytthheedjeett.ecAtilotnhooufgahdoduitrioXn-arlayjedtaktnaoatsr,ewnoethsauvfe-
y
C/S 0.1 shown how the ratio of X-ray to radio intensities for the
ξ I Γ = 20 kXn-roatys oefmciussrivoendpjreotscecsasn. IbCe/uCsMedBasemaidssiiaognn,obsteiicngfomr tohree
tightlybeamedthansynchrotronemissionwouldbeman-
0.01 ifestbyalargerratioforaknotmovingclosertothe line
of sight than its neighbors.
0 5 10 15 20 25
θ Angle to l.o.s. (deg)
Fig. 6.— The function ξ describing the effects of the extra
beaming factor on the ratio of IC/CMB to synchrotron emission.
CurvesforfourrepresentativevaluesofΓ(3,6,10,and20;topto
bottom)areshown.
We thank the anonymous referee for useful comments
that led to improvements in the paper. We thank R.
TABLE 5 Morgantiforgivingusher5GHzVLAmapof3C17. F.
X-rayto Radio Flux RatiofortheJetof 3C17 MassaroisgratefultoG.MiglioriandS.Bianchifortheir
suggestions in the Chandra data analysis, A. Siemigi-
Knot ν∗S5GHz X-rayFlux Ratio nowskaforherhelpintheuseoftheChandra CIAOdata
(10−15cgs) (10−15cgs) reductionanalysissoftware,andE.LiuzzoandS.Giacin-
tuccifor their suggestionsaboutthe radiodata analysis.
S3.7 1.45 6.4 4.4
S10.3 1.06 <1.6 <1.5 ThisresearchhasmadeuseofNASA’sAstrophysicsData
S11.3 3.83 7.1 1.9 System; SAOImage DS9, developed by the Smithsonian
S10.8 3.24 1.6 0.5 AstrophysicalObservatory;andtheNASA/IPACExtra-
S9.6 1.80 <1.6 <0.9 galactic Database (NED) which is operated by the Jet
S8.0 1.25 <1.6 <1.3 Propulsion Laboratory, California Institute of Technol-
S7.2 4.63 <1.6 <0.3 ogy, under contract with the National Aeronautics and
Notestotable Space Administration. The National Radio Astronomy
TheX-rayfluxisfortheband0.5to7keVfromour8ks ObservatoryisoperatedbyAssociatedUniversities,Inc.,
observation. Dividingcol.3bycol.2yieldstheratiosofcolumn4.
under contract with the National Science Foundation.
TheX-rayfluxofS10.8comesfromasingleeventandisreported
hereforillustrativepurposeonly TheworkatSAOissupportedbyNASA-GRANT GO8-
9114A.
Facilities: VLA, HST, CXO (ACIS)
the location of maximum curvature of the jet (Miller &
Brandt 2008).
REFERENCES
Cardelli,J.A.,Clayton,G.C.,Mathis,J.S.1989,ApJ,345,245 Morganti, R., Oosterloo, T., Tadhunter, C. N. 1999 A&AS, 140,
Buttiglione,S.,Capetti,A.,Celotti,A.,Axon,D.J.,Chiaberge,M., 355
Macchetto, F.D.,Sparks,W.B.2008,A&A,submitted Protheroe,R.J.,2004APh,21,415
Dickson,R.,D.1997,PhDthesis,Univ.Sheffield Schmidt,M.1965ApJ,141,1
Donzelli, C. J., Chiaberge, M., Macchetto, F. D. 2007, ApJ, 667, Schombert, J. M.,Pildis,R. A.,Eder, J. A., & Oemler,A., 1995,
780 AJ110,2067
Harris,D.E.,Krawczynski,H.2002,ApJ,565,244 Schwartz,D.A.,etal.2007,Ap&SS,311,341
Harris,D.E.,Krawczynski,H.2006,ARA&A,44,463 Siebert,J.,Brinkmann,W.,Morganti,R.1996,MNRAS,279,1331
Jester, S., Harris,D. E., Marshall, H. L., Meisenheimer, K. 2006, Spergel,D.N.,etal.2007,ApJS,170,377
ApJ,648,900 Tadhunter,C.N.,etal.,1998,MNRAS,298,1035
Kalberla,P.M.W.,Burton, W.B.,Hartmann, D. 2005, A&A, 440, Tavecchio,F.,Maraschi,L.,Sambruna,R.M.2000ApJ,544L,23
775 Venturi,T.,Morganti,R.,Tzioumis,T.,Reynolds,J.2000, A&A,
Ledlow,M.J.,Owen,F.N.1996,IAUS,175,238L 363,84
Miller,B.&Brandt,N.,2008ApJ,submitted
Morganti, R., Killeen,N. E. B., Tadhunter, C. N.,1993 MNRAS,
263,1023