Astron.Nachr./AN999,No.88,789–793(2006)/DOIpleasesetDOI! High energy emission from AGN cocoons in clusters of galaxies M.Kino1,2,⋆ N.Kawakatu2,H.Ito3andH.Nagai2 1 ISAS/JAXA,3-1-1Yoshinodai,229-8510Sagamihara,Japan 2 NationalAstronomicalObservatoryofJapan,181-8588Mitaka,Japan 3 DepartmentofScienceandEngineering,WasedaUniversity,Tokyo169-8555,Japan 9 0 Received30May2005,accepted11Nov2005 0 Publishedonlinelater 2 n Keywords jets–galaxies:active–galaxies:gamma-rays–theory a J Gamma-rayemissionfromcocoonsofyoungradiogalaxiesispredicted.Consideringtheprocessofadiabaticinjectionof theshockdissipationenergyandmassoftherelativisticjetintothecocoon,wefindthatthethermalelectrontemperature 0 ofthecocoonistypicallypredictedtobeoftheorderof∼MeV,andisdeterminedonlybythebulkLorentzfactorofthe 2 jet.Togetherwiththetime-dependentdynamicsofthecocoonexpansion,wefindthatyoungcocoonscanyieldthermal Bremsstrahlungemissionsatenergies∼MeV.Hottercocoons(i.e.,GeV)foryoungersourcesarealsodiscussed. ] A (cid:13)c 2006WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim G . h 1 Introduction AmongavarietyofAGNbubbles,apopulationsocalled p - compactsymmetricobjects(CSOs)hasbeenwidelyinves- o Relativisticjetsinactivegalacticnuclei(AGNs)arewidely tigated in various ways (e.g., Fanti et al. 1995; Readhead r t believedtobethedissipationofkineticenergyofrelativistic et al. 1996; O’Dea & Baum 1997; de Vries et al. 1997; s a motionwithaLorentzfactoroforder∼10producedatthe O’DeaandBaum1998;Stanghellinietal.1998;Snellenet [ vicinity of a super-massive black hole at the galactic cen- al.2000;Dallacasaetal.2000;Girolettietal.2003;Nagai ter(Begelman,BlandfordandRees1984forreviews).The etal.2006;Orientietal.2007;Kawakatuetal.2008).CSOs 1 v jetinpowerfulradioloudAGNs(i.e.,FRIIradiosources) aresmallerthan1kpcandthepreviousstudiessupportthe 8 issloweddownviastrongterminalshockswhichareiden- youth scenario in which CSOs propagate from pc scales 6 tified as hot spots. The shocked plasma then expand side- thrusting away an ambientmedium and growingup to FR 9 waysandenvelopethewholejetsystemandthisissocalled II radio galaxies.CSOs are thusrecognizedas newly born 2 a cocoonor a bubble (Fig. 1). The cocoonis a by-product AGNjets,andtheyarecrucialsourcestoexplorephysicsof . 1 oftheinteractionbetweenAGNjetsandsurroundingintra- AGNbubblesintheirearlydays. 0 clustermedium(ICM).Theinternalenergyoftheshocked In this study, we propose that “young AGN bubbles” 9 plasmacontinuouslyinflatesthiscocoon.Sofarlittleatten- areanewpopulationofγ-rayemittersintheUniverse.The 0 : tion has been paid to observational feature of the cocoon, layout of the paper is as follows. We review the expand- v sincetheyareusuallyinvisibleinGHzbandsbecauseofthe ingcocoonmodelin§2followingourpreviousworks(Kino i X synchrotron cooling for older electrons. 1 As a result, we andKawakatu2005;Kino,KawakatuandIto2007,KKI07 r justseeapartofthecocoon.Thevisiblepartisso-calledra- hereafter) We show the predicted MeV gamma emission a diolobesinwhichrelativelyfreshelectronsarefilledin.In from youngcocoonsin §3 . In §4, we furtherpredictGeV Fig.2,weshowonegoodsampleofcocoonemissionfrom gamma emission for smaller CSOs. Summary and discus- thepowerfulradiogalaxyCygnusAfortheevidenceofits sionisgivenin§5. existence.2 2 Cocooninflationby exhausted jet ⋆ Correspondingauthor:e-mail:[email protected] 1 Onthecontrary,theemissionfromtheshellmadeoftheshockedICM Here we consider the time-evolution of an expanding co- (seeFig.1)hasbeenexploredbymanyauthors(e.g.,Heinz,Reynoldsand cooninflatedbythedissipationenergyoftherelativisticjet Begelman1998;SutherlandandBicknell2007).Sincetheshellshavenon- viaterminalshocks.Theadiabaticenergyinjectionintothe relativisticvelocities,theemissionfromtheshellsispredictedintheX-ray cocoonisassumed.Massandenergyconservationfromthe band. 2 ThedataobtainedbyCarillietal.(1991)wasre-analyzedtoobtainthe jet into the cocoon, which govern the cocoon pressure Pc mapshowninFig.2.TheobservationwascarriedoutwithVLAAconfig- andmassdensityρ arewrittenas c urationat330MHzon1987August18.Thedataanalyzesisperformedby γˆ P (t)V (t) standardmannerwithAstronomicalImageProcessingSystem(AIPS).The c c c ≈2T01(t)A (t) (1) γˆ −1 t j j flux-scaleisdeterminedbycomparisonwith3C286and3C48usingthe c AIPStaskSETJY.TheimageisobtainedwiththeDIFMAPafteranumber ρ (t)V (t) c c ≈2J(t)A(t), (2) ofself-calibrationiterations. j j t (cid:13)c 2006WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim 790 M.Kino,N.Kawakatu,H.Ito&H.Nagai:Highenergyemissionfromcocoons entrainmentoftheenvironmentalmattertakesplaceduring thejetpropagation(e.g.,Mizutaetal.2004).Accordingto this,themassandkineticenergyfluxofthejetareregarded vc as constantin time. Then, the conditionsof T01 = const, j headAh radio lobe andJj =constleadstotherelationsofρj(t)Aj(t)=const andΓ (t)=const.InordertoevaluateL,weusetheshock j j R cocoon jump condition of Γ2ρ = β2 ρ (Kawakatu and Kino v c j j hs ICM h 2006)whereβ (= v /c)andρ isthe advancespeed hs hs ICM jet A Z of the hot spot βhs = 10−2β−2 and the mass density of c hs ICM, respectively. Using, the jump condition, L is given hot spot j by d eICxtMernρρρρal ICM Lj =2×1045Rk2pcβ−22n−2ergs−1 (3) where we use A(t) = πR2 (t), and the hot spot radius j hs R isgivenbyR = R (107 yr)/1 kpc.Asa fiducial hs kpc hs case,wesetthenumberdensityofthesurroundingICMas nICM(d) = ρICM(d)/mp = 10−2 cm−3n−2(d/30kpc)−2 Fig.1 AcartoonrepresentationofinteractionoftheICM wheredisthedistancefromthecenterofICMandcocoon with declining atmosphere and the relativistic jet in FR II (see Fig. 1). Since the change of the index from −2 does radio galaxy.As a result, most of the kinetic energyof jet notchangetheessentialphysicsdiscussedinthiswork,we is deposited in the cocoon and it is inflated by its internal focus on this case for simplicity. Since L is the ultimate j energy. sourceofthephenomenaassociatedwiththecocoon,allof theemissionpowerswhichwillappearin§3shouldbeless thanL. j The number density of total electrons in the cocoon is governed by the cocoon geometry and its plasma content. Forconvenience,we definethe ratioof“thevolumeswept bytheunshockedrelativisticjet”to“thevolumeoftheco- coon” as A(t). We denote V (t) = 2(π/3)R2Z3 (t), Z c hs hs satisfiesZ (t) = β ct,R ,andR ≡ R /Z < 1asthe hs hs c c hs cocoonvolume,thedistancefromthecentralenginetothe hot spot, is the radius of the cocoon body,and the aspect- ratio ofthe cocoon,respectively(e.g.,Kino andKawakatu 2005).PostulatingthatRandZ /R areconstantintime, hs hs A(t)≡ 2Aj(t)vjt isevaluatedas Fig.2 VLA image of Cygnus A at 330 MHz Vc(t) (A-configuration). The contour level starts with A(t)≈0.4R−2Rk2pcZ3−02β−−21 (4) 0.565 Jy beam−1 and increases from there by factors where Z = Z (107 yr)/30 kpc. Note that, in the case, 30 hs of 2. The convolving beam is 4.73 × 3.91 arcsec at the the time dependence of A is deleted since V ∝ t−3 and c position angle of 10.1◦. In this frequency, we can see the A ∝ t2. This case satisfies v = const (e.g., Conway j hs synchrotron emission from slightly colder electrons than 2002).The cocoonmass density ρ (t) is controlledby the c those emitting GHz ranges. Hence we can identify the massinjectionbythejetanditcanbeexpressedasρ (t)≈ c cocoon profile which are clearly different from familiar Γ ρ(t)A=β2 Γ−1ρ (Z (t))Awhereweusetheshock j j hs j ICM hs “classicaldoublelobe”. condition of Γ2ρ = β2 ρ . Adopting typical quanti- j j hs ICM ties of FR II sources(e.g., Begelman, Blandfordand Rees 1984), the number density of the total electrons in the co- whereγˆ ,V ,T01,J andA,aretheadiabaticindexofthe c c j j j coonisgivenby plasmainthecocoon,thevolumeofthecocoon,thekinetic −2 eonfetrhgeyjaent,drmesapsescfltiuvxeloy.fTthheejetot,taalndkitnheeticcroesnse-rsgeyctiaonndalmaraesas ne(t)≈4×10−5A¯n−2Γ10β−22(cid:18)107tyr(cid:19) cm−3(5) flux of the jet are Tj01 = ρjc2Γ2jvj, Jj = ρjΓjvj where ρj, where Γ = 10Γ , andA¯ = A/0.4. Here we assume that 10 and Γj are mass density and bulk Lorentz factor of the jet themassdensityofthee± pairplasmaisheavierthanthat (Blandford and Rees 1974). Hereafter we set v = c. The j of electron-protonone,andthenwe adoptρ ≈ 2m n in c e e totalkineticpoweroftherelativisticjetisdefinedasL ≡ j the light of previous works (Reynolds et al. 1996; Wardle 2T01(t)A (t)anditisassumedtobeconstantintime. j j etal.1998;SikoraandMadejski2000;KinoandTakahara Asforthemassandkineticenergyfluxofpowerfulrela- 2004).Theupperlimitofthermaln canbebasicallycon- e tivisticjets,numericalsimulationstellusthatnosignificant strained by the analysis of Faraday depolarization(Dreher (cid:13)c 2006WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.an-journal.org Astron.Nachr./AN(2006) 791 etal.1987).However,thestrongFaradaydepolarizationob- ∼ 103 times larger than that νF ∼ 10−11 ergcm−2s−1. ν servedinCSOs(Cottonetal.2003)arelikelytobecaused This is still less than the threshold of INTEGRAL. This bydenseforegroundmattersuchasnarrowlineregion.There- may be the reason for the lack of detection of MeV emis- fore n in radio lobes of CSOs has not been clearly con- sion from youngcocoonsup to now.Fig. 3 showsthat the e strained. XMM/Newton satellites can detect the low energy part of Letusestimate theelectron(andpositron)temperature the thermalBremsstrahlung from youngcocoons. In MeV (T ) and proton temperature (T ). From Eqs. (1) and (2) energyband,aproposedmissionofdetectorSGDonboard e p together with the equation of state P ≈ 2n kT , we can the NeXT satellite with the eye up to ∼ 0.6 MeV (Taka- c e e directlyderivethetemperaturesas hashi et al. 2004) could detect the thermal MeV emission from those located slightly closer or youngerwith smaller kT ≈1Γ MeV, kT ≈2Γ GeV (6) e 10 p 10 Lorentzfactor. where we adopt the two temperaturescondition of kT ≈ e (m /m )kT . It should be stressed that the temperatures e p p aregovernedonlybyΓ.Itisalsoworthnotingthatthegeo- j metricalfactorsinEqs.(1)and(2)arecompletelycancelled 3.2 Candidatesources out.Onecannaturallyunderstandthesepropertiesbycom- paring the well-established properties such as supernovae In order to explore the extended cocoon emission in the and GRBs. Constant temperature in AGN jet can be real- X-ray band, one may think it is hard to distinguish over- ized bythe“continuous”energyinjectionintotheexpand- lapping emission from the compactcore of the AGN with ing cocoon whilst temperatures of astrophysical explosive limitedspacialangularresolutionofthecurrentX-raysatel- sourcessuchasgamma-rayburstsandsupernovaewouldbe lites. However, the averaged spectral index in X-ray band decreased because of “impulsive” injection of the energy. (Γ ) from the compact core of AGNs is softer than the Thus the resultant temperatures are uniquely governed by X Bremsstrahlungemission(KoratkarandBlaes1999).Hence Γ andtheyremaintobeconstantintime. j itispossiblefindcandidatesourcesoftheMeVcocoonby the value of Γ . As far as we know, there are two possi- X 3 MeV γ emissionsfrom ayoung cocoon ble candidatesforthe Bremsstrahlungemission.Thoseare B1358+624(Vinkatal.2006)andPKSB1345+125(Siemigi- 3.1 ThermalMeVBremsstrahlungemission nowskaet al. 2008)actuallyshowsΓ ≈ 1,both ofthem X were observed by XMM/Newton. Time variability of ob- Thetime-dependenceofthethermalBremsstrahlunglumi- servedspectraisalsothekeytodistinguishthem.Itisobvi- nosity LBrem is given by LBrem(t) ∝ n2e(t)Te3/2Vc(t) ∝ ousthatthecocoonemissionisconstantintimewhilstvari- t−1 based on the cocoonexpansionshown in the previous ousemissionsfromthecoreofAGNshouldbehighlyvari- section.Henceitisclearthatayoungercocoonsarebrighter able.Hencesteadyemissionsareconvincinglyoriginatedin Bremsstrahlung emitters than older cocoons. In a similar cocoons. way,brightersynchrotronluminosityhasbeenexpectedfor younger radio galaxies (e.g., Readhead et al. 1996). With Intriguingly,“the diffuse X-ray emission” has been in- relativisticthermalBremsstrahlungemissivity(Rybickiand deed detected in PKS B1345+125with the size of the ex- Lightman 1979), the luminosity of the optically thin ther- tendedemissionisoforder∼ 20kpcbySiemiginowskaet malBremsstrahlungemissionνL atenergies∼ 1MeVis al. (2008). The diffuse emission might be associated with ν estimatedas the radiolobesofthis source.TheX-rayemission is elon- −1 gatedtowardstheSouth-WestsimilarlytotheVLBIjetaxis t L (t)≈2×1040n¯2R2Θ3/2 ergs−1.(7)reportedbyStanghellinietal.(2001).Iftheemissionisas- Brem e 10 (cid:18)107yr(cid:19) sociatedwiththeradioemittingplasma,therearetwopos- Eq. (7) explains the reasons for the non-detection of the sibilitiestoexplaintheemission.Herewenewlystressthat thermal emission from older cocoons. One is simply be- the tail of the Bremsstrahlung emission could explain the causeitisnotverybright.Theotherisbecausethepredicted emission. Based on Eq. (7), the X-ray luminosity of the energyrangeis∼1MeV,theMeV-γastronomyisstillim- emissionshowsL ≈1×1043ergs−1.TheobservedL X X matureanditissometimescalledas“sensitivitygap”com- can be explained with the n ≈ 1 × 10−2 cm−3. Based e paredwiththeenergyrangebelow10keVandaboveGeV on Eq. (5), the n can be realized with β ∼ 10−1 and e hs ranges(Takahashietal.2004). n ∼ 10−1 cm−3 forinstance.Sincethe requiredn is ICM e In Fig. 3, we show the predictedvaluesof νF for the considerably large, the analysis of Faraday depolarization ν cocoons with t = 107 yr and t = 104 yr located at the willbecrucialforcheckingtheupperlimitofn .Theother e distance of D = 102 Mpc. The cocoon with t = 107 yr possibility is non-thermalemissions from the lobes which have νF ∼ 10−14 ergcm−2s−1. The detection thresh- is recently investigated by Stawarz et al. (2008).Observa- ν old of SPI instrument on board the INTEGRAL satellite tionsnearthepeakofthermalemissionNeXTsatellitewill is about νF ∼ 10−9 ergcm−2s−1 at ∼ 1 MeV. For a becrucialtodistinguishwhethertheemissionisthermalor ν youngcocoonwitht = 104 yr,thepredictedluminosityis nonthermalone. www.an-journal.org (cid:13)c 2006WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim 792 M.Kino,N.Kawakatu,H.Ito&H.Nagai:Highenergyemissionfromcocoons 4 Younger radio sources as “hotter” bubbles -8 10 keV MeV GeV -9 So far, we discuss the cocoon property in the phase of no 10 Integral significantcooling.Thephaseroughlycorrespondstomedium 10-10 sciozoelisnygmtimmeetsrcicaloesbjbeecctosm(MesShOosr)tearnfodrFsmRaIlIlegrasloauxriceess.,Scionocle- -2-1m s]10-11 NeXT GLAST c10-12 ing effects for CSOs are more effective than the case for g young cocoon largeronessuchasMSOsandFRIIs.Weconsistentlysolve [erν10-13 (t=104 yr) F asetofequationswhichdescribesyoungbubbleexpansions ν 10-14 includingtheeffectsofBremsstrahlungemissionandadia- 10-15 Newton old cocoon batic loss together with the initial conditions indicated by (t=107 yr) -16 CSOobservations.Thenwefindthatthebubbleshaveelec- 10 tron temperature of ∼ GeV at initial phases, the bubbles 102 103 104 105 106 107 108 109 thencooleddowntoMeVbytheadiabaticloss.Wefurther Energy [eV] estimate these γ-ray emissions and show that it could be Fig.3 Model prediction of MeV-peaked thermal detectedwithFermi(GLAST)(Kinoetal.2009). bremsstrahlungemissionfromcocoonslocatedatD =102 Mpc.Thepredictedemissionfromyoungcocoonisbrighter enoughtodetectinX-raybandwhilstthatfromanoldco- 5 Summary anddiscussion coonismuchdarkerthanthedetectionlimits. WehaveinvestigatedtheluminosityevolutionsofAGNco- coons together with the dynamicalevolution of expanding Fanti, C., Fanti, R., Dallacasa, D., Schilizzi, R.T., et al. 1995, cocoon.Belowwesummarizethemainresultsofthepresent A&A,302,317 work. GirolettiM.,GiovanniniG.,TaylorG.B.,ConwayJ.E.,LaraL., 1. We newlypredictthe Bremsstrahlungemissionpeaked VenturiT.,2003,A&A,399,889 Guainazzi,M.,Siemiginowska,A.,Stanghellini,C.,Grandi,P.,Pi- atMeV-γbandasaresultofstandardshockdissipation concelli,E.,AzubikeUgwoke,C.,2006,A&A,446,87 ofrelativisticjetsinAGNs.Thetemperatureoftheco- Gugliucci, N.E., Taylor, G.B., Peck, A.B., Giroletti, M. 2005, coonisgovernedonlybythebulkLorentzfactorofthe ApJ,622,136 jetΓ .TheelectrontemperatureT relevanttoobserved j e Heinz,S.,Reynolds,C.S.,Begelman,M.C.,1998,ApJ,501,126 emissionsistypicallypredictedintherangeofMeVfor Kawakatu,N.,Kino,M.,2006,MNRAS,370,1513 Γ ∼ 10. Constant temperatures of plasma in the co- Kawakatu,N.,Nagai,H.,Kino,M.,2008,ApJ,687,141 j cooncan be realized becauseof the continuousenergy Kino,M.,Takahara,F.,2004,MNRAS,349,336 injectionbythejetwithconstantΓ (KKI07). Kino,M.,Kawakatu,N.,2005,MNRAS,364,659 j Kino, M., Kawakatu, N., Ito, H., 2007, MNRAS, 376, 1630 2. We further investigate younger bubble expansions in- (KKI07) cludingtheeffectsofBremsstrahlungemissionandadi- Kino,M.,Ito,H.,Kawakatu,N.,Nagai,H.,2009,MNRAS,sub- abaticlosstogetherwiththeinitialconditionsofCSOs. mitted(arXiv:0812.1850) Thenwefindthatthelobesinitiallyhaveelectrontem- KoratkarA.,BlaesO.,1999,PASP,111,1 peratureofGeV andthelobesthencooldownto MeV Mizuta,A.,Yamada,S.,Takabe,H.,2004,ApJ,606,804 bytheadiabaticloss.Theγ-rayemissionscouldbede- Nagai,H.,Inoue,M.,Asada,K.,Kameno,S.,Doi,A.,2006,ApJ, tectedwithFermi(GLAST)(Kinoetal.2009). 648,148 O’Dea,C.P.,Baum,S.A.,1997,AJ,113,148 O’Dea, C. P., Mu, B., Worrall, D. M., Kastner, J., Baum, S., de Acknowledgements. WewouldliketothankC.R.KaiserandM. Vries,W.H.,2006,ApJ,653,1115 Sikorafor valuable comments. NK issupported byGrant-in-Aid Orienti,M.,Dallacasa,D.,Stanghellini,C.2007,A&A,475,813 forJSPSFellows.HIacknowledgetheGrantforSpecialResearch Polatidis,A.G.,Conway,J.E.2003,PASA,20,69 ProjectsatWasedaUniversity. Readhead,A.C.S.,Taylor,G.B.,Pearson,T.J.,Wilkinson,P.N., 1996,ApJ,460,634 References Reynolds,C.S.,Fabian,A.C.,Celotti,A.,Rees,M.J.,1996,MN- RAS,283,873 Begelman,M.C.,Blandford,R.D.,Rees,M.J.,1984,Rev.Mod. Siemiginowska, A., LaMassa, S., Aldcroft, T. L., Bechtold, J., Phys.,56,255 Elvis,M.,2008,ApJ,684,811 Blandford,R.D.,Rees,M.J.,1974,MNRAS,169,395 Snellen,I.A.G.,Schilizzi,R.T.,Miley,G.K.etal.2000,MNRAS, Carilli,C.L.,Perley,R.A.,Dreher,J.W.,Leahy,J.P.,1991,ApJ, 319,445 383,554 StanghelliniC.,O’DeaC.P.,DallacasaD.,BaumS.A.,FantiR., Cotton,W.D.,etal.,2003,PASA,20,12 FantiC.,1998,A&AS,131,303 Dallacasa, D., Stanghellini, C., Centonza, M, Fanti, R. 2000, Stanghellini,C.,Dallacasa,D.,O’Dea,C.P.,Baum,S.A.,Fanti, A&A,363,887 R.,Fanti,C.,2001,A&A,377,377 DeVries,W.H.,Barthel,P.D.,O’DeaC.P.,1997,A&A,321,105 Stawarz, Ł., Ostorero, L., Begelman, M. C., Moderski, R., Dreher,J.W.,Carilli,C.L.,Perley,R.A.,1987,ApJ,316,611 Kataoka,J.,WagnerS.,2008,ApJ,680,911 (cid:13)c 2006WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim www.an-journal.org Astron.Nachr./AN(2006) 793 Sutherland,R.S.,Bicknell,G.V.,2007,ApJS,173,37 Takahashi,T.,etal.,2004,NewAR,48,269 Vink, J., Snellen, I., Mack, K.-H., Schilizzi, R., 2006, MNRAS, 367,928 Wardle, J. F. C., Homan, D. C., Ojha, R., Roberts, D. H., 1998, Nature,395,457 www.an-journal.org (cid:13)c 2006WILEY-VCHVerlagGmbH&Co.KGaA,Weinheim