Hydrodynamical simulations of the jet in the symbiotic star MWC 560 III. Application to X-r

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

Received2007February28;accepted2007April04

APreprinttypesetusingLTEXstyleemulateapjv.10/09/06

HYDRODYNAMICALSIMULATIONSOFTHEJETINTHESYMBIOTICSTARMWC560

III.APPLICATIONTOX-RAYJETSINSYMBIOTICSTARS

MatthiasStuteandRaghvendraSahai

JetPropulsionLaboratory,CaliforniaInstituteofTechnology,4800OakGroveDrive,Pasadena,CA91109,USA

Received2007February28;accepted2007April04

arXiv:0704.2240v1 [astro-ph] 17 Apr 2007

ABSTRACT

InpapersIandIIinthisseries,wepresentedhydrodynamicalsimulationsofjetmodelswithparametersrepresentativeofthesymbioticsystemMWC560.Theseweresimulationsofapulsed,initiallyunderdensejetinahighdensityambientmedium.Sincethepulsedemissionofthejetcreatesinternalshocksandsincethejetvelocityisveryhigh,thejetbowshockandtheinternalshocksareheatedtohightemperaturesandshouldthereforeemitX-rayradiation.Inthispaper,weinvestigateindetailtheX-raypropertiesofthejetsinourmodels.WehavefocusedourstudyonthetotalX-rayluminosityanditstemporalvariability,theresultingspectraandthespatialdistributionoftheemission.TemperatureanddensitymapsfromourhydrodynamicalsimulationswithradiativecoolingpresentedinthesecondpaperareusedtogetherwithemissivitiescalculatedwiththeatomicdatabaseATOMDB.ThejetsinourmodelsshowextendedandvariableX-rayemissionwhichcanbecharacterizedasasumofhotandwarmcomponentswithtemperaturesthatareconsistentwithobservationsofCHCygandRAqr.TheX-rayspectraofourmodeljetsshowemissionlinefeatureswhichcorrespondtoobservedfeaturesinthespectraofCHCyg.Theinnermostpartsofourpulsedjetsshowironlineemissioninthe6.4–6.7keVrangewhichmayexplainsuchemissionfromthecentralsourceinRAqr.WeconcludethatMWC560shouldbedetectablewithChandraorXMM-Newton,andsuchX-rayobservationswillprovidecrucialforunderstandingjetsinsymbioticstars.

Subjectheadings:circumstellarmatter—hydrodynamics—ISM:jetsandout ows–binaries:sym-biotic–methods:numerical—X-rays:ISM

1.INTRODUCTION

Highlycollimatedfastout owsorjetsarecommoninmanyastrophysicalobjectsofdi erentsizesandmasses:activegalacticnuclei(AGN),X-raybinaries(XRBs),youngstellarobjects(YSO),pre-planetaryneb-ulae(PPN),supersoftX-raysourcesandsymbioticstars.Inthelasttwoobjects,thejetengineconsistsofanac-cretingwhitedwarf.Insymbioticstars,thecompanionisaredgiantundergoingstrongmassloss.Morethanonehundredsymbioticstarsareknown,butonlyabouttensystemsshowthepresenceofjets.ThemostfamoussystemsareRAquarii,CHCygni,andMWC560.

RAquarii,withadistanceofabout200pc,isoneofthenearestsymbioticstarsandawellknownjetsource.Thejethasbeenextensivelyobservedintheopticalandatra-diowavelengths(e.g.Solf&Ulrich1985;Itshowsarichmorphol-ogy,e.g.aseriesofparallelfeaturesinthejetandthecounter-jet,extendingtoafewhundredAUeach.Thisisahintofpulsedejectionofbothjets.Furthermore,RAqristhe rstjetinasymbioticsystem,whichwasde-tectedinX-raysuKelloggetal.2001).Kelloggetal.(2001)foundpeaksofOVIIat0.57keVinboththeNEandSWjetsandapeakofNVIat0.43keVonlyintheNEjet.ThespectraareconsistentwithasoftcomponentwithkT～0.25keV.ThecentralsourceshowsasupersoftblackbodyemissionwithkT～0.18keVandaFeKαlineat6.4keVwhichsuggeststhepresenceofahardsourcenearthehotstar.Recently,reportedon veyearsofobservationswithChandraandwereable

Electronicaddress:Matthias.Stute@jpl.nasa.gov

tomeasurethepropermotionofintheNEjetofabout600kms 1.investigatedtheX-rayemissionfromtheinner500AUofthissystem.In1984/85,CHCygnishowedastrongradiooutburst,duringwhichadouble-sidedjetwithmultiplecompo-nentswasejectedThisevental-lowedanaccuratemeasurementofthejetvelocitynear1500kms 1.InHSTobservations2002),arcscanbedetectedthatalsocouldbeproducedbyepisodicejectionevents.X-raywas rstde-tectedbyEXOSAT(Leahy&Taylorandsubse-quentASCAobservationsrevealedacomplexX-rayspec-trumwithtwosoftcomponents(kT=0.2and0.7keV)associatedwiththejet,(7.3keV)andaFeKαlineTheyinter-pretedthehardcomponentasthermalemissionbyma-terialbeingaccretedontothewhitedwarfandthesoftcomponentaseithercoronalemissionfromthegiantstaroremissionsfromshocksinthejets.AnalysisofarchivalChandraACISdatabyGalloway&Sokoloski(2004)re-vealedfaintextendedemissiontothesouth,alignedwiththeopticalandradiojetsseeninHSTandVLAob-servations.Wheatley&KallmanreanalyzedtheASCAdataandinterpretedthesoftemissionasscatter-ingofthehardcomponentinaphoto-ionizedmediumsurroundingthewhitedwarf.Theyclaimthatnoothersourcesthantheaccretingwhitedwarfarerequiredtoexplainthespectrum.Theobviousexistenceofthejet,however,andfurthermoretheapparentdeclineofthehardX-raycomponentobservedwiththeUS-JapaneseX-raysatelliteSuzakubyMukaietal.togetherwiththelackofacorrespondingdeclineinthesoftcom-ponent,suggestthatthisinterpretationisimplausible.

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

2

Recently,Karovskaetal.(2007,hereafterKCR07)re-portedthedetectionofmultiplecomponents,includinganarc,inthearchivalChandraimages.

RAqrandCHCygaretheonlytwojetsofsymbioticstarswhicharedetectedinX-rays.Whilethesetwoob-jectsareseenathighinclinations,inMWC560thejetaxisispracticallyparalleltothelineofsight.Thisspe-cialorientationallowsonetoobservetheout owinggasaslineabsorptioninthesourcespectrum.Withsuchobservationstheradialvelocityandthecolumndensityoftheout owingjetgasclosetothesourcehasbeeninvestigatedingreatdetail.Inparticulartheacceler-ationandevolutionofindividualout owcomponentsandjetpulseshasbeenprobedwithspectroscopicmon-itoringprograms,asdescribedinSchmidetal.(2001).Usingthisopticaldata,sophisticatednumericalmod-elsofthispulsedpropagatingjethavebeendeveloped(Stute,Camenzind&Schmid2005;Stute2006,here-afterPaperIandIIinthisseries).Anumberofhydro-dynamicalsimulations(withandwithoutcooling)weremadeinwhichthejetdensityandvelocityduringthepulseswerevaried.Thebasicmodelabsorptionlinepro lesinMWC560aswellasthemeanvelocityandvelocity-widthareingoodagreementwiththeobserva-tions.Theevolutionofthetime-varyinghighvelocityabsorptionline-componentsisalsowellmodeled.Thesemodelsnotonly ttheMWC560data,butarealsoabletoexplainpropertiesofjetsinothersymbioticsystemssuchastheobservedvelocityandtemperatureoftheCHCygjet.

Sofar,MWC560hasnotbeendetectedinX-rays(M¨ursetetal.1997).We ndusingthePIMMStool1thatthenon-detectionintheROSATall-skysurveysetsan1upperlimitoftheabsorbedX-ray uxof0.07countss (M¨ursetetal.1997)and7×10 13ergs 1cm 2,respectively.

Thejetsinallthreesymbioticstarsshowevidenceofepisodicity.Suchepisodicityintheejectionprocesshasbeenseeninnumericalmodelsoftheinteractionofthestellarmagnetosphereandtheaccretiondisk(e.g.Goodsonetal.1997;Mattetal.2002).Inthedisk-windscenario(e.g.Blandford&Payne1982;Andersonetal.2003)thetime-dependentemissioniscreatedbytimevariationsintheaccretionrateoftheunderlyingdisk.Unfortunately,sofarnohydrodynamicalmodelsexistforexplainingtheX-rayemissionfromsymbioticstars.Asa rststepwehavethereforeusedourexistingsim-ulations,which tMWC560,forunderstandingtheob-servedX-rayemissionpropertiesofMWC560,CHCygandRAqr.

In§2,webrie ydescribethenumericalmodelswehaveused.ThetotalX-rayluminosityanditstimedepen-denceisexaminedin§3.Aftertheresultingspectraarecalculatedin§4,weshowemissionmapsin§5andap-plyourmainresultstoX-rayobservationsofCHCyg,MWC560andRAqrin§6.Finallyourconclusionsaregivenin§7.

2.THENUMERICALMODELS2.1.Thehydrodynamicalsimulations

1

http://heasarc.gsfc.nasa.gov/Tools/w3pimms.html

Wesolvedtheequationsofidealhydrodynamicswithanadditionalcoolingtermintheenergyequation

ρ

t

+ (ρv v)= p ρ Φ e

CPwhichwaswrittenby

Ziegler&Yorke(1997)andmodi edbyThiele(2000)tocalculateradiativelossesduetonon-equilibriumcool-ingbylineemission.ρisthegasdensity,pthepressure,etheinternalenergydensity,Φthegravitationalpoten-tial,vthevelocityandγtheratioofthespeci cheatsatconstantpressureandvolumewhichissettoγ=5/3.ThegeneralcapabilitiesofthecodehavebeendescribedindetailinPaperI,forourapproximationsandassump-tionsrelatedtothecoolingtreatmentwereferthereadertoPaperII.Webrie ydescribethegeometrywhichwehaveadoptedinoursimulationsbelow.

Weuseacylindricalcoordinatesystemwherethejetaxiscorrespondstothezaxisandbothcomponentsofthebinarysystemarelocatedintheplaneperpendiculartothisaxis.Thehotcomponentislocatedattheoriginofthecoordinateframe;withabinaryseparationof4AU,aredgiantisimplemented.Theredgiantissurroundedbyastellarwindwithconstantvelocityof10kms 1andamasslossrateof10 6M⊙yr 1.

Thejetisproducedwithinathinjetnozzlewitharadiusof1AU.Theinitialvelocityofthesteadyjetischosento1000kms 1anditsdensityissetto8.4×10 18gcm 33(equaltoahydrogennumberdensityof5×106cm ). These8parametersleadtoi)amasslossrateof≈10M⊙yr 1inthesteadyjet,andii)adensitycontrastbetweenthesteadyjetandtheambientmediumηof5×10 3andaMachnumberof≈60inthejetnozzleattheoriginofthecoordinatesystem.Repeatedlyeachseventhday,thevelocityanddensityvaluesinthenozzlearechangedtosimulatethejetpulseswhichareseenintheobservationsofMWC560.Thedurationofeachpulseisoneday.

Twomodels(iv’andi’)outofourexistingsetofeightmodelswerechosenforcomputingX-rayemissionprop-erties.Thesemodelsrepresentmaximum(modeliv’)andminimum(modeli’)valuesofthejetdensityinthepulses.Inmodeliv’(modeli’),thejetdensityinthepulsesishigher(lower)thanthejetdensityinthesteadyjet.Althoughmodeliv’providedthebest tfortheop-ticaldataforMWC560,ourworkinthispapershowsthatmodeli’resultsinX-raypropertieswhicharemoreappropriateforCHCyg.Forbothmodelsweusedanapproximatetreatmentofradiativecooling.ThemodelpropertiesincludingthevelocitiesanddensitiesofthejetpulsesaregiveninTable1.

Mapsoflogarithmofdensityandtemperatureformodeli’onday162andformodeliv’onday115aregiveninFig.1.Botharethelasttime-stepscalculated.

2.2.CalculatingtheX-rayproperties

WedeterminedtheexpectedX-ray uxusingtheden-sityandtemperaturemapsfromthehydrodynamical

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

3

TABLE1

Parametersofthejetpulses

modeli’iv’

npulse[cm 3]1.25×1061×107

vpulse[kms 1]

20002000

˙jp[M⊙yr 1]M

4.66×10 93.73×10 8

Ljet[ergs 1]5.88×10334.70×1034

˙js=9.33×10 9M⊙yr 1andLjet=2.93×1033Note.—Thevaluesofthesteadyjetemissionaren=5×106cm 3,v=1000kms 1,M

˙jsandM˙jparethemassout owratesofthejetinthesteadystateandduringthepulse,respectively.Thedurationergs 1,respectively.M

ofeachpulseisoneday,theirperiodisseven

days.

Fig.1.—Logarithmofdensity(top)andtemperature(bottom)formodeli’onday162(left)andformodeliv’onday115(right),shownascontourplotsandslicesalongthejetaxis(solid)andparalleltothejetaxisatr=2AU(dashed).

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

4

Fig.2.—X-rayluminosityasafunctionoftheevolutiontimeofrangethei.e.0.15jetin–model15i’(solid)andmodeliv’(dashed)intheenergykeVtheareplotssimilar.

showingkeV.Mosttheluminosityoftheenergyintheisemittedenergyrangebelow0.152keV,–2simulationsasfollows.WeusedtheatomicdatabaseATOMDBwithIDLincludingtheAstrophysicalPlasmaEmissionDatabase(APED)andthespectralmodelsoutput

fromtheAstrophysicalPlasmaEmissionCode(APEC,Smithetal.2001)tocalculatetheemissivity.ThedefaultabundancesinATOMDB,i.e.14elements(H,He,C,N,O,Ne,Mg,Al,Si,S,Ar,Ca,Fe,Ni)withsolarabundancesofAnders&Grevesse(1989),areused.Theenergyrangeisdividedintobinsof0.01keV.Wecomputethespectrumandthetotal uxinX-raysasafunctionofevolutionarytimeforeachofourmodels.WecalculatetheX-rayemissionintherangebetween0.15–15keV,whichisexactlytheenergyrangecoveredbyEPIConXMM-NewtonandincludesthatoftheACISinstrument(0.2–10keV)andofHETG(0.4–10keV)onChandra.Theemissionfromgaswithatemperaturelowerthan 106Kisonlymarginalinthisenergyrange.

3.THETOTALX-RAYLUMINOSITYANDITSTIME

DEPENDENCE

Asexpected,thehightemperatures,createdbythein-teractionofthejetpulseswithpreviouslyejectedmatter,leadtosubstantialX-rayemission(Fig.2).TheX-rayluminosityinmodeliv’ishigherthaninmodeli’.ThisisaresultofthehigherdensityinthepulsesofthehigherkineticluminosityL˙andthus

v2pumpedjet.Furthermore,inmodeljet=1/2M

intothei’about5%oftheaveragekineticluminosityisradiatedinX-rays,butinmodeliv’about19%.SincetheX-raysareemittedbyshockedmaterialfromthefastmovingpulsesandsincetheX-rayluminosityisproportionaltoρ2,comparedtoLjetbeingproportionaltoρ,theratiooftheX-raylu-minositytothekineticluminosityisproportionaltoρ.Thereforethisratioishigherinmodeliv’thaninmodeli’.

We ndminimaandmaximaintheX-rayemissionL(computedbyintegratingovertheenergyrange0.15–15XkeV)whichareconnectedwiththeperiodicemergenceofjetpulses(Fig.3).ThustheperiodofthevariationsintheX-rayemissionisabout7days.Thesizeofthe uctuationsis50%andmoreoftheaverageemission.WhiletheX-rayluminositystaysconstantwithtimeformodeli’,itdecreaseswithtimeformodeliv’.Thisdi erencemightberelatedtoalargeramountofcoolinginmodeliv’.Theinitialshocktemperatureisidenticalinbothmodels,sincethevelocitiesarethesame.The

Fig.3.—X-rayluminosityasafunctionoftimeformodeli’(top)createdandmodeliv’(bottom);onecanindecreasesthetopbyplotstheemergenceshowthedi erentofeachnewtrendsjetseeofpulse;minimatheX-raytheanddottedmaximaluminositylineshigherdensitieswithtime.

inthejetpulsesinmodeliv’,however,leadtohigherdensitiesintheX-rayemittingmaterialandthustohigherpressureswhichresultinstrongeradi-abaticexpansionandhenceenhancedadiabaticcooling.Radiativecoolingisalsoenhancedbythehigherdensitiesinmodeliv’.

4.THESPECTRUMANDITSTIMEDEPENDENCE

Thespectraofbothmodelsintheenergyrangebe-tween0.15–15keVshowmanydi erentfeatures.Theyshowcontinuumemission,andsuperimposedonthecon-tinuum,alargenumberofemissionfeatures(someofwhichareblendsofseveralemissionlines).Aprominentfeature,whichismainlyduetoblendedironlines,isseenbetween0.7–1keV.Ironalsoproducesastrongemissionfeatureinthe6.4–76.7keVrangerequiringveryhightemperatures(～10K)thatarereachedlocallyinthejet.

LikethetotalX-rayluminosity,thespectrumisalsohighlytime-dependent(Fig.4).Wede netwoproxiesforthetemperature,oneusingthelowenergyspectrumandoneusingthehighenergyspectrum.Theseproxiescanthenbeusedconvenientlyfordirectcomparisonwiththesingle-temperaturethermalplasmamodelstypicallyusedto ttheobserveddata.

4.1.De ningtemperatureproxies

Inordertocharacterizethetemperatureoftheprop-agatingjetfromthelowenergyspectrum,weusethefactthatbelowenergiesofabout0.7keV,bothspectrainFig.4arealmostidentical,butdi ersigni cantlybe-tween0.7and2keV.Thereforewede netheproxyζfor

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

Fig.4.—Spectrumintheenergyrangebetween0.15–15keV;top:minosity,formodelity,solid)i’andondays155(close153(atominimumthemaximumoftheX-raytotalX-raylu-(dashed).dashed),wardsseethebytimeaThebottom:factorspectrumfordependenceof100plottedmodeliv’ondays105(solid)luminos-and107offortheclaritywithspectrum.

inaeachsolidplot.lineisOneshiftedcanclearlydown-relativelylowtemperatureplasma(107K)inthejetas

ζ=

f(0.2 0.7)keV

f(3)

9keV

Wechooseregionsinthespectrumwherenolinesarepresent,althoughphotonswiththesehighenergieshavenotbeenobservedfromthejetormightbeconfusedwithphotonsfromthecentralengineinobservedspectra.

4.2.Determiningthetemperaturefrommodelspectra

Inmodeli’,thehightemperatureproxy,log(ξ),varieswithtimespanningarangefrom1.5to3.Ithasitshigh-estvalue(i.e.thespectrumhasthesteepestslope,hencetheaveragetemperatureisatitslowest

value),whentheX-rayluminosityshowsaminimum(Fig.6).Compar-isonofthemodellog(ξ)valueswiththatofasingle-temperaturethermalplasma(Fig.5,bottom)givesustemperatureestimatesofthehotcomponentbetween8×106K(0.69keV)and1.7×107K(1.5keV).Thehighesttemperaturesareconsistentwiththatofpost-shockgaswithashockvelocityofabout1100kms 1.Theminimuminthelog(ξ)(i.e.maximuminaverage

5

Fig.5.—Temperatureproxiesasafunctionoftemperatureforahardnesssingle-temperaturetext.

ratioζ(top)thermaland uxplasmaratioξcalculated(bottom)aswithde nedATOMDB:intheFig.6.—X-rayluminosityofthejetasafunctionoftimeintheenergy(middle)range0.15–15keV(top), uxratioξasafunctionoftimemodelsi’.

andhardnessratioζasafunctionoftime(bottom)fortemperature)coincideswiththeemergenceofeachnewpulse;withinthenext2–3dayperiodthecompressedknotcoolsandtheemissivityincreases.ThereforethemaximumintheX-rayluminosityisreachedabout2–3dayslater.

Thelowtemperatureproxy,log(ζ),variesbetweenabout0.5and1.6,thecorrespondingtemperaturesliebetween1.6×106K(0.14keV)and3×106K(0.26keV).Thejetisthereforebetterdescribedasacombi-nationofawarmandahotcomponentratherthanasasingle-temperatureplasma.

Inmodeliv’,log(ξ)variesbetween2.16and2.8;thecorresponding7temperaturesare8×10K(0.69keV)and1.2×10K(1.03keV).Thelowtemperatureproxy,log(ζ),liesbetween0.1and0.6;thecorrespondingtem-peraturesare3×106K(0.26keV)and3.8×106K(0.33keV).Asinmodeli’,thejetisbettercharacterizedasacombinationofawarmandahotcomponent.

Therangeoftemperaturesinthehotcomponentoveronepulsecycleinmodeliv’issmallercomparedtothatinmodeli’;thisisbecausethehigherdensityinmodeliv’makesradiativecoolingmoree cient,suchthattheshockheatingisdampedmoree ciently.Thedi erentdensitycontrastsbetweenthejetpulsesandthesteadyjetinbothmodelsalsoleadtodi erentshockvelocitiesandthustodi erentshocktemperaturestowhichtheplasmaisheatedinitially.

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

6

Fig.7.—sameasFig.

6,butformodeliv’

Fig.8.—Spectrumintheenergyrangebetween0.15–15keVformodelveryusingdi erenti’ondaystwocomponentsionization155;severalatpotentialsstrongemissiondi erenttemperatures.

whichcanlinesonlyarebepresentexplainedwith4.3.Emissionlinesinthemodelspectra

Wehaveidenti edthestrongestemissionlineschar-acterizingatypicaljetspectrumusingthespectrumde-rivedfrommodeli’onday155asatemplate(Fig.8).Themostprominentlinesaretheoxygenlineat0.57keV,theNelinesat0.93and1.03keV,theMglinesat1.35and1.47keV,theSilinesat1.86,2.01and2.18keV,andtheFecomplexatabout6.5keV.Allstronglineswiththeiridenti cationsaregiveninTable2,in-cludingtheir uxesatdays153and155inmodeli’anddays105and107inmodeliv’.MostoftheselinesarehydrogenicorHe-likelinesofheavyelements,however,alsolinesofhigherionizationstatesarepresent(FeXXII,FeXXIII,FeXXV).Thusthissetoflineswithverydif-ferentionizationpotentialscanalsoonlyexplainedwithtwocomponentsatdi erenttemperatures.

5.THEX-RAYEMISSIONMAPS

Theresultsintheprevioussectionsuggesttheexis-tenceoftwocomponents,awarmonewithtemperaturesintherangeof(1.6–3.8)×106Kandahotonewithtemperaturesof(8–17)×106K,respectively.AsalreadypointedoutinPaperIandII,thejetconsistsofdense,coolknotsandtenuous,hotinter-knotgas.TheknotsaretoocoldtoemitX-rays,thusthelowandhighenergycomponentsintheX-rayspectrumprobethetempera-turestructureofthehotpartsofthejet(Fig.1andFig.12inpaperI).Thesehotpartsconsistofshockedmate-rialintheinter-knotgassegments.Temperaturegradi-

entsarepresentwithineachinter-knotgassegmentandalsobetweentheinter-knotgasclosetothejetsourceandthosedownstreamnearthejethead.Theregionswiththehighesttemperaturesandthusemissionabove6keVlieinthe rsttwointer-knotgassegmentswithinonlyafewAUfromthecentralsource(Fig.9).Theinter-knotgassegmentsbecomeprogressivelycooler,astheymovefartherawayfromthejetsourceduetoadiabaticexpansionandtheresultingcoolingofjetmaterial.

PARISONWITHOBSERVATIONS

WenowcomparetheresultsofourmodelstoX-rayobservationsofthethreeobjectsMWC560(onlyupperlimitstothe uxareavailable),CHCygandRAqr.

6.1.MWC560

Thesource uxesinthe0.2–2.4keVrangeare3×10 13ergs 1cm 2formodeli’and2×10 12ergs 1cm 2formodeliv’,usingadistanceof2.5kpctoMWC560(Schmidetal.2001,andreferencestherein).The 13latter uxisreducedtoanabsorbed uxof1.7×10ergs 1cm 2,usingthemodelofthevisualinterstellarextinctionintheGalaxybyHakkilaetal.(1997) 2whichgivesanAforMWC560.Thev=0.88orNabsorbed uxisH=1.55×1021cmconsistentwithMWC560’snon-detectionintheROSATall-skysurvey(<7×10 13ergs 1cm 2,M¨ursetetal.1997).The uxes,however,arehighenoughsuchthattoday’sX-raytelescopeasChandraandXMM-NewtonshouldbeabletodetectMWC5602.

6.2.CHCyg

SincethejetvelocitiesaresimilarinMWC560andCHCyg,wecanmakeadetailedcomparisonofourmodelswiththelatterobject.ThiscomparisonislimitedtothesoftemissionfromthepropagatingjetandexcludestheX-rayemissionaboveabout2keVwhichisbelievedtobedominatedbythevariablescatteredhardX-raysfromthecentralsource.

Ezukaetal.(1998)resolvedforthe rsttimeatomicemissionlines(ortheirblends)fromelementsinhydro-genicandHe-likeionizationstatesintheX-rayspectraofCHCyg.Themostprominentfeaturesintheobservedspectraaretheoxygenlineat0.57keV,theblendofi)Nelinesat0.93and1.03keV,ii)Mglinesat1.35and1.47keV,iii)Silinesat1.86,2.01and2.18keV,andtheFelinecomplexatabout6.5keV.Theselinesarealsoseeninourmodelspectra.Furthermore,theyalsohadtointroduceatwo-temperaturethermalplasmamodeltoexplainthesetoflinesdetectedintheASCAspec-trum.Thetemperaturesoftheirtwocomponents(0.21and0.72keV)areinthesamerangeasinourmodeli’(warmcomponent:0.14–0.26keV,hotcomponent:0.69–1.5keV,see§4.2).Inmodeliv’,thewarmcomponentistoohot(0.26–0.33keV)toexplaintheobservations.Recently,KCR07reportedthedetectionofmultiplecomponents,includinganarc,inthearchivalChandraimages.ThisarchasasimilaropeningangletomanyofthearcsvisibleinourX-rayemissionmaps(Fig.9),anditspresencesupportsourpulsed-jetmodelinwhich

2

executedXMM-NewtoninAO-6afterobservationsMay2007.

proposedbytheauthorswillbe

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

7

TABLE2

Prominentemissionlinesinthemodelspectra

ion

energy(keV)

ux(ergs 1cm 2)

modeli’modeliv’

day153day155day105day

107

Fig.9.—Emissionmapsofmodeli’onday155inthe0.15–1.5keVrange,inthe1.5–3keVrange,inthe3–6keVrangeandinthe

6–7keVrange.

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

8

suchemissionresultsfrominternalshocksgeneratedbycollidingjetejecta.

Consideringthesmallerdistanceof268pctoCHCyg,themodelX-ray33luminosityof2.2×1032ergs 1formodeli’and1.5×10ergs 1formodel 11iv’inthe0.2–2keVrangeimplies uxesof22.6×10ergs 1cm 2and1.74×10 10ergs 1cm ,respectively.TheinterstellarextinctionforCHCygisAv=0.006orNH=1×1020cm 2,thustheabsorbed uxis2×10 11ergs 1cm 2formodeli’and1.3×10 10ergs 1cm 2formodeliv’,respectively.Themeasured uxesofthesoftcompo-nentsassociatedwiththejet,however,lieintherange(0.4 3.8)×10 12ergs 1cm 2(Ezukaetal.1998;Galloway&Sokoloski2004;Mukaietal.2006).Sincethesemeasured uxesaresmallerthanthosepredictedbymodeli’andbyfarsmallerthanthosefrommodeliv’whichhasahigherjetpulsedensitythanmodeli’,weinferthatthejetpulsedensitiesinCHCygaresmallerthanthoseinMWC560andourmodels.If,asstatedin§3,theratiooftheX-rayluminositytothekineticen-ergypumpedintothejetisproportionaltothejetpulsedensity,wecanestimatethatthejetpulsedensityhastobereducedtoabout105cm 3tomodeltheX-ray uxobservedinCHCyg.Inordertoresolvethedi erencebetweenmeasuredandmodel uxeswithuncertaintiesinthedistancetoCHCyg,thelatterwouldhavetobeabout700pc,however,thislargevalueisveryunlikely,sincethedistancewasmeasuredwithHIPPARCOSwithanerrorof23%.

KCR07giveanobservationalestimateofthedensityinthejet,50cm 3,basedonthetotalX-rayluminosity,assuminganemittingspherewitharadiusof400AUandameanemissivityof2×10 23ergcm3s 1.How-ever,astrophysicaljets,typically,arecollimatedstruc-tureswhichhavesigni cantlysmallervolumesthanasphere.Sincewesimulatedthepropagationofthemodeljetsonlyuptoalengthof65AU,not400AU,wescaleupitsvolumeasfollows.Ifweassumethatthecross-sectionalareastaysconstantasthejetpropagatesandevolveswithtime,thejetvolumeisabout(π×62×400)AU3(Fig.9)whichissmallerbyaboutafactorof6000thanthatofaspherewitharadiusof400AU.Alter-natelyifweassumethatthelateralexpansionofthejetisproportionaltoitsaxialexpansion3,thevolumeisstilloverestimatedbyafactorofabout160.Furthermore,withinthevolumeofthejet,theX-raysareemittedbyclumpsandnotbythewholejet,i.e.the llingfactorfortheX-rayemissionis,ftemperatureof1.21<1.Inaddition,KCR07assumea×107K(1keV)toestimatetheemissivity,however,ourjetmodelsshowthatsuchhightemperaturesareonlyarchievedintheinnermostregionofthejet;atlargerdistancesfromthejetsourcealongtheaxis,theX-rayemittingknotsaresigni cantlycooler(Fig.1).ThusthevalueoftheemissitivityshouldbelowerandthedensityshouldbehigherthanestimatedbyKCR07byanothercorrectionfactorf2>1.HenceweconcludethatanaccurateestimateoftheX-rayemit-tingvolumeandthetemperaturewouldleadtoahigherdensitybyafactorof(160–6000)fgivenbyKCR07,oftheorder2/f1comparedtothevalueof104–106cm 3.ThisrangecompareswellwiththedensitiesofmaterialemittingsoftX-raysinourmodelswhichareoftheorder

3

whichisseeninourmodelsafterday70(PaperII)

of105–106cm 3(Fig.1).

OtherpossibilitiestoreducetheX-ray uxesinourmodels,bringingthemclosertotheobservedonesinCHCyg,arei)alongertimescalebetweenthepulsesandii)asmallervelocitydi erencebetweenthesteadyjetandthejetpulses.Inthe rstcase,lessenergyispumpedintothejetandeachjetknotandthejetheadcancoolfurtherbeforebeinghitbythenextpulse.Inthesecondcase,thesmallervelocitydi erencereducesthetemperaturetowhichtheshockedmaterialisheated.However,thedensitycontrastthenalsohastobeadjustedinordertoreproducetheobservedpropermotionofthejetknots.Whichoftheabovescenariosisthemostlikelyonecanonlybedeterminedfromfuturesimulationswhichhavebeen ne-tunedto tthepropertiesoftheX-rayemittingmaterialinthejetofCHCyg.

In1994and2006,the uxfromthejetinCHCygwasalmost 1atthesamelevel,intherange(2.7 3.8)×10 12ergscm 2(Ezukaetal.1998;Mukaietal.2006).In2001,however,itwaslowerbyafactoroften(Galloway&Sokoloski2004).Inthecontextofourmod-els,suchadropin uxmayresultfromalargedecreaseinthedensityofthepulseswhichmaybecausedbyadropintheaccretionrateontothewhitedwarf.

6.3.RAqr

SincethejetvelocityinRAqrissmallerbyafactorof2thanthoseinourmodels,wecanonlymakeamorelimitedcomparisonofthemodelresultswiththeob-servations.Kelloggetal.(2006)andalsoKorrecketal.(2006)reportatangentialmotionofanX-rayemittingknot4of600kms 1.Korrecketal.(2006)estimatedadensityof100cm 3,however,itmaybepossiblethat,asinthecaseofCHCyg,theX-rayemittingvolumeisoverestimatedandthusthedensityisunderestimated.Kelloggetal.(2001)foundinRAqrthattheNEjetismoreluminousbyafactorof3thantheSWjet.ThespectrumtheNEjetwas ttedwithasingle-temperaturethermalplasmawithatemperatureof1.66keV,thespec-trumoftheSWjetwithaplasmatemperatureof0.2keV.Thesimplestexplanationofthisdi erenceisthattheambientmediaonbothsideshavedi erentdensitieswhichwouldleadtodi erentcompressionfactorsandthustodi erentshockheatingtemperatures.However,propermotionmeasurementsshownosigni cantdi er-encesinthevelocitiesoftheknotsinbothsidesofthejet(Paresce&Hack1994;M¨akinenetal.2004),whichwewouldexpectifdi erentambientdensitieswouldde-celeratethejetdi erentlyonbothsides.

Ourmodelingcanprovideaplausibleexplanationfortheobserveddi erencesbetweenthetwosidesofthejet,ifweassumethattheejectionofthejetpulsesonbothsidesareoutofphasewitheachother.Hollisetal.(1991)derivedakinematicageofbothjetsofabout90yrs;overtheextentofthejetwecanobserve3–6knotsintheob-servationsofe.g.Paresce&Hack(1994)whichsuggestsseveralejectioneventsinthisperiodoftime.Aperiodofabout17yearsfortheseejectioneventshasbeenin-ferredfromradioobservations(Kafatosetal.1989),thustheyaresigni cantlylargerthaninMWC560andinour

4

1Korrecketal.(2006)giveanevensmallervelocityof380kmsfrominthetheirvalueTablegiven1,inhowever,thetext.

withoutexplainingthedi erence

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

models5.WehypothesizethattheX-rayemittingblobsintheNEjetwereejectedlaterthanthoseintheSWjetandthereforehavecooledless.Thishypothesisissup-portedbythefactthattheX-rayemittingcomponentintheNEjetisclosertothecentralcore(about8”)thantheblobsintheSWjet(12–26”,Kelloggetal.2001).AnewSWjetcomponentwithano setofabout1.5”fromthecentralsourcehasrecentlybeenreportedbyNicholset al.1(2007).Assumingajetvelocityofabout600kms(Kelloggetal.2006),i.e.about0.6”peryear,weobtainakinematicageofabout2.5yearsforthiscomponent.EvenifthenewcomponenthadbeenejectedduringtheepochofKelloggetal.(2001)’sobser-vations,its uxwouldhavecontributedtothecentralsource,butnottothatoftheSWjet.Hencethepres-enceofthenewcomponentdoesnotcon ictwithourhypothesis.ThetimeperiodbetweentheemergenceofthenewSWjetcomponentandtheemergenceoftheblobnowlocatedat12”isabout17yearswhichsupportstheinferredperiodfortheejectionofjetpulsesinRAqr.In§3,weshowedthatthetotalX-rayluminosityde-creaseswithtime,probablyduetoadiabaticcoolinginthejet(PaperII).Thise ectprovidesaplausibleex-planationofthedecreaseinX-ray uxinthejetofRAqrfromavalueof5×10 13ergs 1cm 2intheearly1990s(H¨unschetal.1998)to1×10 13ergs 1cm 2in2000(Kelloggetal.2001).Ifthisexplanationholds,weexpecttoseeafurtherdecreasein uxinfutureobser-vations.

6.4.The6.4–6.7keVironlinecomplex

Thisironlinecomplexhasbeenobservedinbothob-jects,CHCygandRAqr.OurmodelspectraalsoshowtheexistenceofthisFelinecomplex(§4).Ezukaetal.(1998) ttedtheobservedspectrumofCHCygwiththreesingle-temperaturecomponents(awarmandahotcom-ponenttoexplainthejetemissionandhardcomponentforthecentralengine)andanadditionalGaussianrep-resenting uorescenceemissionintheFeKαline.This uorescenceoccursclosetothewhitedwarfandtheac-cretiondisk.Sinceitisnotpossibletodisentanglethe uxofthethermalandthe uorescencecomponentsinthislinecomplexandsinceourmodelsdonotincludethee ectof uorescence,wecannotcompareourmodelswiththispartoftheobservedspectrumofCHCyg.InRAqr,theoriginofthehardX-rayemissionismoreambiguous.Kelloggetal.(2001)detected16photonsintherangebetween6.4and6.7keVwhichtheyattributetothecentralsourceduetotheextractionregionstheychose.Thephysicaloriginofthisemission,i.e.ther-malor uorescence,cannotbedecided,sincetherearenotenoughphotonstomodelthespectruminthisen-ergyandcharacterizeitsnature.Inourmodel,we ndsigni canthardradiationincludingcontinuumandironemissionlines,beingemittedbythe rsttwointernalshocksinthejetdownstreamfromthesource(i.e.atadistancelessthan15AU).SinceatRAqr’sdistanceof200pc,15AUcorrespondto75milliarcseconds,whichiswellbelowtheangularresolutionofChandra,theX-rayemissionfromtheseshockscannotbeseparatedfromthe

5

passage,TheeventsinRAqrarethoughttodiskinstabilities.

whilethevariationsinMWC560betriggeredseemtobebyaperiastronresultof9

centralsource.ThemodelX-ray24luminosityinthe6.4–6.7keV1rangeisbetween5.1×10ergs 1and7.3×1028ergs formodeli’andbetween2.8×1028ergs 1and5.1×1029ergs 1formodeliv’,respectively,depend-ingonwhetherthejetisinitsminimumormaximumstate.Atadistanceof200pc,this2correspondsto uxesbetween1.1×10 18ergs 1cm and1.1×10 13ergs 1cm 2 .14SinceKelloggetal.(2001)measureda uxof4.9×10ergs 1cm 2at6.41keV,wesuggestthatthemeasuredironline uxmaybeemittedbythejetitself,andanadditional uorescencecomponentisnotneeded,inRAqr.However,newsimulationswiththesamejetvelocityasinRAqrareneededinordertotestthissuggestion.

7.CONCLUSIONS

Wehaveusedourmodelsofpulsed,radiativejetsinsymbioticstarsinordertoinvestigatetheirX-rayprop-ertiesindetail.Thesemodelsshowthatthewell-studiedpole-onjetinMWC560shouldbeeasilydetectedbytoday’sX-raytelescopessuchasChandraandXMM-Newton,sinceourmodel uxand13itstimevariationforthissourceareoftheorderof10 ergs 1cm 2.

We ndminimaandmaximaintheX-rayemissionL(computedbyintegratingovertheenergyrange0.15–15XkeV)whichareconnectedwiththeperiodicemergenceofjetpulses.ThemaximaofthetotalX-rayluminos-ityoccur2–3daysaftertheemergenceofnewjetpulses,whichareejectedevery7days.Thesizeofthe uctua-tionsis50%andmoreoftheaverageemission,makingsuchX-ray ashingjetsdetectablewithChandraandXMM-Newton.

TheX-rayspectraofourmodeljetsarerichinemissionlinefeatures,themostprominentofwhichcorrespondtoobservedfeaturesinthespectraofCHCyg.

Byusinglowandhighenergytemperatureproxiesde-rivedfromthespectra,wecanshowthattheemissioncanbeadequatelycharacterizedwithahotandawarmoptically-thinplasmacomponent.Thehotcomponenthastemperaturevaluesofabout0.7keV(1.6keV)duringtheminima(maxima)ofthetotalX-rayluminosityandthewarmcomponenthastemperaturevaluesofabout0.14keV(0.33keV)duringtheminima(maxima).

Whilemodeliv’isappropriateforMWC560,wehaveshownthatmodeli’,whichhasalowerjetpulsedensitythanmodeliv’,ismoreappropriateforthejetinCHCygintermsofexplainingthelowerX-ray ux.Otherpossi-bilitiestoreducethe uxarealongertimescalebetweenthepulsesandasmallervelocitydi erencebetweenthesteadyjetandthejetpulses.Whichoftheabovesce-nariosisthemostlikelyonehastobetestedinfuturesimulationswhichhavetobetunedtothejetinCHCyg.Ourmodelsalsoo eraplausibleexplanationforthedi erencesinluminositiesandtemperaturesintheNEandtheSWjetofRAqr.Weassumethattheejectionofthejetpulsesonbothsidesareoutofphasewitheachother.WehypothesizethattheX-rayemittingblobsintheNEjetwereejectedlaterthanthoseintheSWjetandthereforehavecooledless.

We ndtheexistenceofironlineemissioninthe6.4–6.7keVrangeinourmodelswhichisalsoobservedinboth,CHCygandRAqr.OurmodelscannotbedirectlyappliedtoCHCyg,becauseofanadditional uorescencecomponentfromthecentralsourceandaccretiondiskin

In papers I and II in this series, we presented hydrodynamical simulations of jet models with parameters representative of the symbiotic system MWC 560. These were simulations of a pulsed, initially underdense jet in a high density ambient medium. Since th

10

theobservedspectrumandbecause uorescenceisnotin-cludedinourmodels.InthecaseofRAqr,althoughthisemissionhasbeenassociatedwiththecentralsource,ourmodelingshowsthatitisconsistentwithbeingproducedbyjetknotsveryclosetothelatter,becausetheirsepa-rationinourmodeliswellbelowtheangularresolutionofChandra.

Usingourcurrentmodelswhichwerebuiltto ttheopticaldataofthejetinMWC560weareabletoexplainsomeoftheimportantcharacteristicsofX-rayemissionfromjetsinMWC560andothersymbioticstars.Theresultsofthisstudydemonstratethegreatpotentialoffuturenumericalsimulationsofpulsedjetswhichhavebeen ne-tunedtospeci csourcepropertiesforunder-standingthejetphenomenoninsymbioticstars.Fur-Anders,E.,Grevesse,N.1989,GeCoA53,197

Anderson,J.M.,Li,Z.-Y.,Krasnopolsky,R.,Blandford,R.D.2003,ApJ590,L107

Blandford,R.D.,Payne,D.G.1982,MNRAS199,883Ezuka,H.,Ishida,M.,Makino,F.1998,ApJ499,388

Eyres,S.P.S.,Bode,M.F.,Skopal,A.,etal.2002,MNRAS335,526l

Galloway,D.K.,Sokoloski,J.L.2004,ApJ613,L61Goodson,A.P.,Winglee,R.M.,B¨ohm,K.-H.1997,ApJ489,199Hakkila,J.,Myers,J.M.,Stidham,B.J.,Hartmann,D.H.1997,AJ114,2043

Hollis,J.M.,Michalitsianos,A.G.,Kafatos,M.,etal.1985,ApJ289,765

Hollis,J.M.,Lyon,R.G.,Dorband,J.E.,etal.1985,ApJ475,231

Hollis,J.M.,Oliversen,R.J.,Michalitsianos,A.G.,etal.1991,ApJ377,227

Hollis,J.M.,Pedelty,J.A.,Lyon,R.G.1997,ApJ482,L85H¨unsch,M.,Schmitt,J.H.,Schroeder,K.,Zickgraf,F.1998,A&A330,225

Kafatos,M.,Hollis,J.M.,Yusef-Zadeh,F.,etal.1989,ApJ346,991

Karovska,M.,Carilli,C.L.,Raymond,J.C.,Mattei,J.A.2007,ApJaccepted,astro-ph/0703278

Kellogg,E.,Pedelty,J.A.,Lyon,R.G.2001,ApJ563,151

Kellogg,E.,Anderson,C.,DePasquale,J.,etal.2006,posterabstract#70.08,AmericanAstronomicalSocietyMeeting207Korreck,K.E.,Kellogg,E.,Sokoloski,J.L.2006,in:”TheMulticolouredLandscapeofCompactObjectsandtheirExplosiveOrigins”,astro-ph/0611401

thermore,wehopethatthisstudywillinspirenewandmoresensitiveobservationsofX-rayemissionfromjetsinsymbioticstars.

Weacknowledgehelpfulandimprovingcommentsandsuggestionsbythereferee.Thehydrodynamicalsimula-tionswereperformedattheHighPerformanceComput-ingCenterStuttgart,Germany.ThisworkwaspartiallyfundedbyNASA/CHANDRAgrantsGO3-4019XandGO4-5163Z,andNASA/STScIgrantHST-GO-10317.01-A.TheresearchdescribedinthispublicationwascarriedoutattheJetPropulsionLaboratory,CaliforniaInsti-tuteofTechnology,underacontractwiththeNationalAeronauticsandSpaceAdministration.

Leahy,D.A.,Taylor,A.R.1987,A&A176,262M¨akinen,K.,Lehto,H.J.,Vainio,R.,Johnson,D.R.H.2004,A&A424,157

Matt,S.,Goodson,A.P.,Winglee,R.M.,B¨ohm,K.-H.2002,ApJ574,232

Mukai,K.,Ishida,M.,Kilbourne,C.,etal.2006,PASJaccepted,astro-ph/0609245M¨urset,U.,Wol ,B.,Jordan,S.1997,A&A319,201

Nichols,J.S.,DePasquale,J.,Kellogg,E.,etal.2007,ApJacceptedParesce,F.,Hack,W.1994,A&A287,154

Schmid,H.M.,Kaufer,A.,Camenzind,M.,etal.2001,A&A377,206

Smith,R.K.,Brickhouse,N.S.,Liedahl,D.A.,Raymond,J.C.2001,ApJ556,L91

Solf,J.,Ulrich,H.1985,A&A148,274Stute,M.,2006,A&A450,645(PaperII)

Stute,M.,Camenzind,M.,Schmid,H.M.2005,A&A429,209(PaperI)

Sutherland,R.S.,Dopita,M.A.1993,ApJS,88,253

Taylor,A.R.,Seaquist,E.R.,Mattei,J.A.1986,Nature,319,38Thiele,M.2000,Numericalsimulationsofprotostellarjets,PhDThesis,UniversityofHeidelberg

Viotti,R.,Piro,L.,Friedjung,M.,Cassatella,A.1987,ApJ319,L7

Wheatley,P.J.,Kallman,T.R.2006,MNRAS372,1602Ziegler,U.,Yorke,H.1997,m.101,54

REFERENCES

本文来源：https://www.bwwdw.com/article/h5cm.html