Astronomy&Astrophysicsmanuscriptno.EtaCarPaper2 (cid:13)c ESO2012 January18,2012 Constraints on the non-thermal emission from η Carinae’s blast wave of 1843 J.L.Skilton1,W.Domainko1,J.A.Hinton2,D.I.Jones1,S.Ohm2,3,andJ.S.Urquhart4 1 Max-Planck-Institutfu¨rKernphysik,POBox103980,D69029Heidelberg,Germany e-mail:[email protected] 2 X-rayandObservationalAstronomyGroup,DepartmentofPhysicsandAstronomy,UniversityofLeicester,LE17RH,UK 3 SchoolofPhysicsandAstronomy,UniversityofLeeds,LS29JP,UK 4 CSIROAstronomyandSpaceScience,P.O.Box76,Epping,NSW1710,Australia 2 1 Received;Accepted 0 2 ABSTRACT n Non-thermalhardX-rayandhigh-energy(HE;1MeV<E<100GeV)γ-rayemissioninthedirectionofηCarinaehasbeenrecently a detectedusingtheINTEGRAL,AGILEandFermisatellites.Thisemissionhasbeeninterpretedeitherintheframeworkofparticle J acceleration inthecolliding windregionbetween thetwomassivestarsor intheveryfast moving blast wavewhichoriginatesin 7 thehistorical1843“GreatEruption”.ArchivalChandradatahasbeenreanalysedtosearchforsignaturesofparticleaccelerationin 1 ηCarinae’sblastwave.Noshell-likestructurecouldbedetectedinhardX-raysandalimithasbeenplacedonthenon-thermalX-ray emissionfromtheshell.ThetimedependenceofthetargetradiationfieldoftheHomunculusisusedtodevelopasinglezonemodel ] E fortheblastwave.AttemptingtoreconciletheX-raylimitwiththeHEγ-rayemissionusingthismodelleadstoaveryhardelectron injectionspectrumdN/dE∝E−ΓwithΓ<1.8,harderthanthecanonicalvalueexpectedfromdiffusiveshockacceleration. H Keywords.Accelerationofparticles,stars:binaries,stars:individual:ηCarinae,gamma-rays:stars,X-rays:stars . h p - o 1. Introduction nificant variability in the 50keV and MeV-GeV regime is sur- r prising in a colliding wind binary (CWB) picture, particularly st η Carinae is a binary system composed of a massive pri- during periastron passage where a collapse of the CWR is ex- a mary (M ≥ 90M⊙, η Car A) and a less massive secondary pected (see e.g. Parkinetal. 2009, and references therein) and [ (M ≤ 30M⊙, η Car B) star (see e.g. Nielsenetal. 2007) hencenoparticleaccelerationshouldoccur.Theblast-wavesce- 1 which orbit each other in 2022.7±1.3days (Daminelietal. narioprovidesagoodexplanationoftheobservedemissionbe- v 2008a). η Carinae experienceda historicaloutburst(the “Great causeoftheexistenceofanextendedemissionregion,thusex- 3 Eruption”)inthe19thcenturyandejected∼12M⊙ ofgaswhich plaining the lack of significant variability of the source. While 3 movesoutwardsatanaveragespeedof∼650kms−1(Smithetal. FermiandINTEGRALdonotprovidesufficientangularresolu- 5 2003) formingthe “HomunculusNebula”.Recent observations tiontoresolvetheblastwave,X-rayobservationswiththecur- 3 show that η Carinae is surrounded at a distance of ∼0.25pc rentgenerationofinstrumentsshouldprovideresolvedimagesof 1. by a very fast moving blast wave (3500 − 6000kms−1) pro- theregion.Howeverthereislikelytobesomeconfusionofthe 0 duced in the giant outburst of 1843 (also known as the “Great possiblenon-thermalX-rayshellwiththermalemissionfromthe 2 Eruption”) (Smith 2008). This blast wave currently overruns Outer Ejecta (see Sewardetal. 2001, for more details). In this 1 the “Outer Ejecta” a ring-likestructure of materialwhich orig- work,archivalChandraX-raydatahavebeenanalysedtosearch v: inates from an ejection of mass from η Carinae ∼ 500−1000 forsignaturesofacceleratedparticlesintheblast-waveregion. yearsago(Walbornetal.1978).Thisfast-movingmaterialmim- i X icsalow-energysupernovaremnant(SNR)shell(Smith2008), r withablastwavemovingintotheISMwithvelocitiescompara- 2. Chandradata a ble tothe historicalsupernovaeRCW86(Vinketal. 2006)and SN1006(Vink2005). An 88.2ks Chandra ACIS-I observation (obs ID 6402) of the Thereisevidenceforthepresenceofrelativisticparticlesin Trumpler16 region,taken in August 2006 was reprocessed us- ηCarinae.Theexistenceofnon-thermalX-rayemissionwasre- ing the most recent version of the calibration files. Data pro- cently reported by the INTEGRAL collaboration (Leyderetal. cessingandreductionwasperformedusingtheChandraCIAO 2008).Inthehighenergy(HE;1MeV< E <300GeV)domain, (v4.3) software package and CALDB (v4.4.3). The data were a source spatially coincident with the η Carinae position was unaffectedbysoft-protonflaresandsothefullobservationtime reported by the AGILE and Fermi collaborations (Tavanietal. was available for analysis. Standard procedureswere followed 2009; Abdoetal. 2009, 2010). Within the measured positional and images (count-maps) were created in two energy bands; uncertainties of INTEGRAL and Fermi-LAT, particle accelera- 0.3keV–5keVand5keV–10keVandareshowninFigure1. tion via the diffusive shock acceleration (DSA) process is pos- Theextendedemissionattributedtotheexpandingejectaaround sible in the colliding wind region (CWR) of η Carinae and/or η Carinae can be seen clearly in the low-energy image. The intheexpandingblastwaveoftheGreatEruptionof1843(see high-energyimagehoweverisdominatedbythebrightpoint-like Ohmetal. 2010,fora detailedmodelling).Theabsenceofsig- emission from η Carinae itself. Several other low-significance 1 Skiltonetal.:Non-thermalemissionfromηCarinae’sblastwave (2010)wasbasedonananalysisofthefirst11monthsofFermi- LAT observations. More recent analyses of increased data sets are now available and are used here to constrain the model of Ohmetal.(2010). An analysis including 21 months of Fermi-LAT data has been presented by Farnieretal. (2011). This analysis reveals two components of the HE emission; a low-energy part (here- after L-component) which is best described by a power-law with spectral index Γ = 1.69 ± 0.12 and exponential cut-off at E = 1.8 ± 0.5GeV, and a second, high-energy compo- c nent(hereafterH-component)extendingto≈ 100GeV,bestde- scribedbyapurepowerlawwithindex1.85±0.25.Notempo- ral variability in either component is reported in Farnieretal. (2011). However, some variability in the H-component is re- ported by Walter&Farnier (2011). Farnieretal. (2011) have proposedthatthe Fermi-H componentresults fromthe interac- tionofacceleratedprotonsandnuclei.Thisinterpretationisat- tractive in the sense that accelerated protonscan have a higher maximum acceleration energy and suffer less from losses than electrons.Wefollowthisapproachinthisdiscussionattributing thetwocomponentstoelectrons(L-component)andprotons(H- component)intheblastwave. 4. OriginoftheHEγ-rayemission ThepropertiesoftheHEγ-rayemissionfromtheregioninand around η Carinae are extremely challengingto interpretwithin theframeworkofanycurrentmodel.Inparticular,formodelsin whichtheemissionoriginatesintheCWR,thelackofobserved variabilityforthebulkoftheemissionishardtoreconcilewith the dramatic changes seen at other wavelengths during perias- tronpassage(seeDaminelietal.2008b,andreferencestherein) andtheveryshortcoolingtimeofrelativisticparticlesinthesys- tem (Farnieretal. 2011; Bednarek&Pabich 2011). The outer Fig.1.Chandraimageofthe regionaroundηCarinae.Thetop blast-wave scenario proposed in Ohmetal. (2010) provides a (bottom)panel shows the countsmap in the energyrange 0.3– promisingalternativeinthesensethatshorttimescalevariability 5.0keV (5.0–10.0keV). The black ellipse and circle show the is not expected. However, the Chandra observations presented fluxextractionregionfortheejecta.Seetextfordetails. here place rather tight constraints on this scenario with impor- tantconsequencesforshockaccelerationinsystemsofthistype. sources are detected above 5keV, see Leyderetal. (2010) for Here we presentthese constraintsand the refined modelof the details. blast wave emission and discuss the more generalimplications Slices in Right Ascension and Declination were made ofourresults. through the low and high energy images and through a simu- latedimageoftheChandra-ACISPSF(createdatthesamechip- 4.1.ImprovedBlastWaveModel locationasηCarinaeinthisobservation).Thesliceregionswere 7′′ wideandtheprojectionsareshowninFigure2.Itisclearto The single-zone, time-dependent numerical model used in seethatthereislittleemissionabovethePSFinthehighenergy Ohmetal.(2010)anddescribedindetailinHinton&Aharonian map(greenpoints)fromtheηCarinaeregion. (2007) has been modified here such that the radiation field en- The flux from the shell was estimated from the elliptical ergy density of the Homunculus nebula, which had been as- region shown in Figure 1. A circular region (also shown in sumed to be static is now modelled as a function of time. The Figure 1) centredon the position of η Carinae with a radiusof HomunculusisilluminatedfromtheinsidebythestarηCarinae 0.305′wasexcludedfromthefluxcalculation.Thebackground withthe(bolometric)luminosityL .Thisstarlightisreprocessed η was estimated from a large elliptical region to the south west bydustintheopticallythicknebulaintoIRlight.Inthesteady- of the “on region”. All point sources were removed from the statecase,itsownluminosityL = L ,butshiftedtotheIRband. H η backgroundregionbeforeextractingthe flux.The background- Itsthermaltimescale,t ,isgivenbyt ≈ E /L ,with E be- th th th H th subtracted flux from the shell in the 5-10keV band was calcu- ing the thermal energy of the Homunculus nebula. For a mass latedtobe4.3×10−13ergcm−2s−1.Nospectralanalysisofthere- oftheHomunculusnebulaof12M andatemperatureof260K ⊙ gion has been attemptedand so this value representsthe upper E (Gehrz&Smith 1999) is foundto be about1045erg.Using th limitonthenon-thermalfluxfromtheejecta. L ≈ 1040ergs−1 (Coxetal.1995)leadstoathermaltimescale H oftheorderofafewweeks.Thisvalueismuchsmallerthanthe ageoftheHomunculusnebulaandtherefore,assumingasteady 3. Fermi-LATdata fluxfromthestarηCarinae,a(pseudo-)steadystateisareason- Thehypothesisthatparticleaccelerationisoccurringintheex- able assumption. Any luminosity change of the star η Carinae panding ejecta surrounding η Carinae suggested by Ohmetal. wouldhencebefollowedbyacorrespondingluminositychange 2 Skiltonetal.:Non-thermalemissionfromηCarinae’sblastwave 4.2.Applicationtothedata unts600 o xcess C500 dFuigc.e3dsbhyoewlesctthreonγs-raacycespleercattreadleinnethrgeybdlaissttriwbauvtieonanwdhiinctheriascptirnog- E withthetime-dependentradiationfieldoftheHomunculusneb- 400 ula. The X-ray upper limits presented in this work restrict the modelconsiderably.Thespectralindexofthe acceleratedelec- 300 tronsisconstrainedbytheChandralimittoberatherhard(Γ < 1.8).Thisvalueismuchharderthanthecanonicalvalue(Γ=2) 200 expectedfromdiffusiveshockacceleration,butcouldberealised inverystrongshocksorinshockswhicharemodifiede.g.bythe 100 pressureoftheacceleratedparticlepopulation.Usingamagnetic fieldstrengthof10µG(asusedinOhmetal.2010),thetotalen- 0 ergyin electrons E for this modelwouldbe 6×1045erg, rep- -0.015 -0.01 -0.005 0 0.005 0.01 0.015 e ∆ Right Ascension (°) resentingonlyaverysmallfraction(≈ 10−4)ofthetotalkinetic energyE intheblastwaveof≈ (4−10)×1049ergwhichisin k principleavailableforparticleacceleration.Foraslightlyhigher unts600 magneticfieldstrengthof20µG,andaspectralindexofΓ=2.1, o xcess C500 ttoheaegnreeergwyiitnhetlheectXro-nrasyhalismtiot.bHeoswmeavlleerr,tthhaisnw3o×ul1d04im5 ipnlyortdheart E the vast majority of the HE γ-ray emission does not originate 400 in the blast wave. Given these findings, the fraction of energy innon-thermalelectronscomparedtothetotalkineticenergyin 300 the blast wave is even lower compared to the model presented before. The association of the soft γ-ray emission detected by 200 INTEGRALfromtheηCarinaeregion(Leyderetal.2008)with the blastwave is problematicdueto the sharp,lowenergycut- 100 off required for consistency with the Chandra limits presented both here and in Leyderetal. (2010). It seems likely that this 0 hard X-ray emission is associated with the CWB as suggested -0.015 -0.01 -0.005 0 0.005 0.01 0.015 ∆ Declination (°) byLeyderetal.(2010)ratherthantheblastwave. Fig.2. Slices through the Chandra images in Right Ascension Ohmetal. (2010) concluded that hadrons were less likely (toppanel)andDeclination(bottompanel)(seetextfordetails). responsible for the single-component HE γ-ray emission re- The width of the extractionregionwas 7′′. Black pointsrepre- vealed in the first 11 months of Fermi-LAT data. This conclu- sent the profile of the full energy range (0.3–10.0keV) image. sionwasderivedfromtwofactors;thepresenceofHEemission Red points represent the low (<5keV) energy data and green at ≈ 200MeV (below the ≈ 300MeV threshold energy for π0 pointsrepresentthehigh(>5keV) energydata. Thebluecurve production)andthefactthatthemaximumenergyofprotonsas showsthesimulatedChandra-ACISPSFatthechip-positionof indicated by the curvature in the Fermi spectrum lay well be- η Carinae in this observation. The dashed vertical lines show lowtheexpectedmaximumaccelerationenergyassociatedwith thesize oftheexcludedregionaroundEtaCarinae(circularre- either the age or size of the system. It has also been argued gion in Figure 1). The y-scale has been truncated to highlight (Farnieretal. 2011; Bednarek&Pabich 2011) that the density thebehaviourawayfromthecentralpeak.NotethatthePSFis oftargetmaterialforppinteractionandsubsequentπ0-decayγ- normalisedtothemaximumofthefull-energy-rangedata. rayproductionwouldbetoolowintheηCarinaeregion. The new Fermi-LAT data reveal the two-component na- of the Homunculus. The temperature of the radiation field T ture of the HE emission, and present a good case that the H- at the location of the (expanding)ejecta is given by the Stefan componentmaybeofhadronicorigin(Walter&Farnier2011). Boltzmannlaw: The variability detected in the H-componentpoint to an origin of at least part of this H-componentflux in the colliding wind 1 T =(cid:16)LH/(4πσR2H)(cid:17)4 region.HoweveritislikelythatsomepartofthisH-component originatesfromprotonsacceleratedintheblastwavesurround- where, R (t) is the time-dependentradius of the Homunculus. ingηCarinae.Suchemissionwouldbenon-variableandwould H Duetothe massoftheexpelledmaterialR (t) followsfreeex- contributetothetotalfluxoftheH-component.Scalingthetotal H pansionR (t)≈v t,withv beingthevelocityoftheejectaof fluxintheH-componentdownbyafactorofthree(representing H H H theHomunculus.Hence,thetimedependentradiationfieldonly thedecreaseinfluxinthehigh-energyFermicomponentfound dependsontheevolutionofthe positionoftheblastwavewith by Walter&Farnier 2011, see green data points in Fig. 3) and respect to η Carinae, the luminosity of η Carinae and the tem- using the available targetmaterialdensity of 100cm−3 leads to peratureoftheradiationfield.Basedonthehistoricallightcurve therequirementof6×1048ergofenergyinhadroniccosmicrays ofHumphreysetal.(1999),andthefactthattheIRemissionis in the region.Thisrepresents6−15%of the kinetic energyof simplyreprocessedstar-light,we assumethattheintegratedlu- theblastwave,giventheuncertaintiesinthekineticenergyesti- minosityofthenebulainIRislinearlyrisingwithEtaCarvisible matesintheblastwaveof4-10x1049erg(Smithetal.2003)– magnitudem . conditionsthatcanreasonablybemet. V 3 Skiltonetal.:Non-thermalemissionfromηCarinae’sblastwave ) 1 -s 2 -m 10-10 Fermi c g INTEGRAL r e ( 10-11 E d / N d 10-12 Chandra 2 E 10-13 10-14 Molonglo 10-15 10-16 10-6 10-4 10-2 1 102 104 106 108 1010 1012 1014 Energy (eV) Fig.3.SpectralenergydistributionfortheregionwithinafewarcminutesofηCarinaeasdescribedinOhmetal.(2010).Curves showasinglezonetimedependentmodelforcontinuousinjectionofelectronsandprotons(synchrotroninblue,inverse-Compton inredandπ0-decayingreen).Amagneticfieldstrengthof B = 10µGandelectronenergyofE = 6×1045ergisassumedforthe e modelrepresentedbytheblueandredsolidcurves.AspectralindexofΓ=1.8andmaximumelectronenergyofE =110GeV max,e isusedforthismodel. E = 6×1048ergofenergyinprotonsandamaximumenergyof E = 2TeVisrequiredtoreproduce p max,p the π0-decayγ-raycomponentindicatedbythe greensolidcurve.Note thatthegreendatapointsare simplythe Fermi-LATdata pointsaspresentedbyFarnieretal.(2011),scaleddownbyafactorofthree.Thedashedsetofcurvesuseamodelwhichassumea magneticfieldofB=20µG,E =3×1045ergofenergyinelectronsandspectralindexΓ=2.1withthesamemaximumelectron e energyofE =110GeVasusedbefore. max,e 5. Summaryandconclusions References We have re-analysed archival Chandra data and placed limits Abdo, A. A., Ackermann, M., Ajello, M., et al. 2010, VizieR Online Data Catalog,2188,80405 on the non-thermal X-ray emission from the expanding ejecta Abdo,A.A.,Ackermann,M.,Ajello,M.,etal.2009,ApJS,183,46 surrounding η Carinae. The single-zone numerical model of Bednarek,W.&Pabich,J.2011,ArXive-prints1104.1275 Ohmetal. (2010) has been adapted to account for the time- Cox,P.,Mezger,P.G.,Sievers,A.,etal.1995,A&A,297,168 varying radiation field and to fit to recent Fermi-LAT HE γ- Damineli,A.,Hillier,D.J.,Corcoran,M.F.,etal.2008a,MNRAS,384,1649 ray data. The two-component nature of the HE emission is Damineli,A.,Hillier,D.J.,Corcoran,M.F.,etal.2008b,MNRAS,384,1649 Farnier,C.,Walter,R.,&Leyder,J.2011,A&A,526,A57+ best explained by electrons (L-component) and protons (H- Gehrz, R. D. & Smith, N. 1999, in Astronomical Society of the Pacific component)respectively.Anattemptto reconcilethenewlimit ConferenceSeries,Vol.179,EtaCarinaeatTheMillennium,ed.J.A.Morse, onthenon-thermalX-rayemissionfromtheshellwiththeFermi R.M.Humphreys,&A.Damineli,251 L-componentdataleadstoaratherhardelectroninjectionindex Hinton,J.A.&Aharonian,F.A.2007,ApJ,657,302 of Γ <1.8. A steady hadronic emission component originating Humphreys,R.M.,Davidson,K.,&Smith,N.1999,PASP,111,1124 Leyder,J.,Walter,R.,&Rauw,G.2008,A&A,477,L29 inthe blastwave,atafluxlevelapproximatelyonethirdofthe Leyder,J.,Walter,R.,&Rauw,G.2010,A&A,524,A59+ totalH-component,canbeexplainedusingreasonablenumbers Nielsen,K.E.,Corcoran,M.F.,Gull,T.R.,etal.2007,ApJ,660,669 fortheenergyincosmicrays,andthetargetdensityintheblast Ohm,S.,Hinton,J.A.,&Domainko,W.2010,ApJ,718,L161 waveregionandthemaximumprotonenergy. Parkin,E.R.,Pittard,J.M.,Corcoran,M.F.,Hamaguchi,K.,&Stevens,I.R. 2009,MNRAS,394,1758 Seward,F.D.,Butt,Y.M.,Karovska,M.,etal.2001,ApJ,553,832 Smith,N.2008,Nature,455,201 Acknowledgements Smith,N.,Gehrz,R.D.,Hinz,P.M.,etal.2003,AJ,125,1458 Tavani,M.,Sabatini,S.,Pian,E.,etal.2009,ApJ,698,L142 SOacknowledgesthesupportoftheHumboldtfoundationbya Vink,J.2005,inAIPCS,Vol.774,X-rayDiagnosticsofAstrophysicalPlasmas: Feodor-Lynenresearchfellowship. Theory,Experiment,andObservation,ed.R.Smith,241–251 4 Skiltonetal.:Non-thermalemissionfromηCarinae’sblastwave Vink,J.,Bleeker,J.,vanderHeyden,K.,etal.2006,ApJ,648,L33 Walborn,N.R.,Blanco,B.M.,&Thackeray,A.D.1978,ApJ,219,498 Walter,R.&Farnier,C.2011,inICRCproceedings 5