/
RadialStellarPulsationsUptothispoint,youhavestudiedstarsinhydrostatice RadialStellarPulsationsUptothispoint,youhavestudiedstarsinhydrostatice

RadialStellarPulsationsUptothispoint,youhavestudiedstarsinhydrostatice - PDF document

phoebe-click
phoebe-click . @phoebe-click
Follow
352 views
Uploaded On 2015-10-02

RadialStellarPulsationsUptothispoint,youhavestudiedstarsinhydrostatice - PPT Presentation

10d4120090291whereistheperiodandthe nal029isthescatteraroundthemeanrelationTheothernumbersareestimatederrorsinthemeanrelationitselfThisformulahasbeentakenfromareviewby5Theslo ID: 147220

10d4:12(0:09)[0:29];(1)whereistheperiod andthe nal[0:29]isthescatteraroundthemeanrelation.Theothernumbersareestimatederrorsinthemeanrelationitself.Thisformulahasbeentakenfromareviewby[5].Theslo

Share:

Link:

Embed:

Download Presentation from below link

Download Pdf The PPT/PDF document "RadialStellarPulsationsUptothispoint,you..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

RadialStellarPulsationsUptothispoint,youhavestudiedstarsinhydrostaticequilibrium.Inthenextfewlectures,wewillconsideroscillatingstars.Webeginwiththesimplestbutperhapsmostimportantcase:radialpulsators.Thesestarsremainsphericallysymmetric,thoughnothydrostatic.1TwoimportantclassesofradialpulsatorsareRRLyraeanddeltaCepheidvariables.(Theterm\variable"canbeappliedtoanystarwhosebrightnessvaries.Inadditiontovarioustypesofpulsators,theseincludeeclipsingbinaries,novae,andotherstarswhosebrightnessvariationshavenothingtodowithpulsation.)RRLyraesandCepheidsareimportantbecausetheyareusedasdistanceestimators.Thesestarsarebrightandcanberecognizedatgreatdistances.Furthermore,theirperiodsarecloselyrelatedtotheirluminosities.Hencebymeasuringtheperiodweknowtheabsolutemagnitudeofsuchastar,andbycomparingthiswiththeapparentmagnitude,wededucethedistance.DeltaCepheidsareevolvedstarswithmasses510M ,usuallyinthehelium-core-burningphase.TheybelongtoPopulationIandarefoundinthedisksofspiralgalaxies.(ThereareothersortsofCepheidvariables,forexamplebetaCepheids.ThedeltaCepheidsarethebrightestandmostuseful,however,sooneusuallycallsthemsimply\Cepheids,"orsometimes\classicalCepheids.")TheirabsolutevisualmagnitudesareMV=1to6(comparetheSun:MV; =+4:83).Theperiod-luminosityrelationforCepheidswasdiscoveredbyHenriettaLeavittin1906([4]),andithasplayedanimportantroleintheGalacticandextragalacticdistancescaleeversince.ArecentdeterminationisMV=2:88(0:20)log 10d4:12(0:09)[0:29];(1)whereistheperiod,andthe nal[0:29]isthescatteraroundthemeanrelation.Theothernumbersareestimatederrorsinthemeanrelationitself.Thisformulahasbeentakenfromareviewby[5].Theslope(thecoecientoflogabove)canbedeterminedfromanygroupofCepheidsatacommonbutunkowndistance(e.g.intheLMC)andisthereforerelativelyuncontroversial.Butthezero-point(givenas4:12above)mustbedeterminedbycalibrationagainstCepheidswhosedistanceisknown,andisstillvigorouslydebated.CepheidsremaincrucialpartofthedistanceladderandhenceofthemeasurementoftheHubbleconstant.RRLyraestarsareoldhorizontal-branchstars(hencealsocoreheliumburning),lowmetallicity,andmasses.1M .TheybelongtoPopulationII;theyarefoundinglobularclustersandintheGalacticbulge.ThehorizontalbranchissonamedbecauseitformsasequenceofroughlyconstantluminosityintheHRdiagram.ThereforeRRLyraescoveramuchsmallerrangeofluminositiesthanCepheids;typically,MV+0:5!+0:2.TheyarefainterthantheCepheids,thoughstillmuchbrighterthantheSun.Withintheirnarrowrangeofmagnitudes,however,theRRLyraesalsosatisfyaperiod-luminosityrelation.ARoughExplanationofthePeriod-LuminosityRelationCepheids,RRLyraes,andmanyotherpulsatorsarefoundinanarrowandnearlyverticalstripintheHRdiagram.Thusthee ectivetemperaturesofthesestarsarecloselysimilarandonlyweaklydependentonluminosity(Te 61103K).Weshallseewhythisistrueshortly.SinceL=4R2T4e andTe constant,L/R2:(2)Theperiodofthefundamentalradialmodeofpulsation,inwhichtheentirestarexpandsandcontractstogether,isdeterminedbythemeandensity:(G)1=2/M R31=2:(3)EliminatingthestellarradiusRbetweentheselasttwoequations,one nds/L3=4M1=2:(4) 1Amoredetailedtreatmentofthissubjectisgiveninxx38-39ofthetextbook[3].1 SincetheclassicalCepheidsarepost-main-sequenceobjects,wecannotusethemain-sequencemass-luminosityrelationtoeliminateMinfavorofL.Nevertheless,sinceLisusuallyasensitivefunctionofmass,massisinsensitivetoluminosity.Thusequation(4)suggests/Lxwithx3=4sincedM=dL�0.However,Te isnotstrictlyconstantwithintheinstabilitystrip:itslowlydecreaseswithincreasingL.Theempiricalrelation(1)indicates/L0:870:06ifoneignoresvariationsinthebolometriccorrection.FrequenciesofAdiabaticRadialModesTheradialmodesofastarcanbelikenedtostandingsoundwavesinanorganpipe.Forourpurposes,anorganpipeisacylindercappedatoneendbutopenattheother.Thepressureandvelocityperturbationsinastandingacousticwaveare90outofphase.Thusattheclosedendofthepipe,thevelocityvanishesandthepressure\ructuationsaremaximal.Similarly,atthecenterofaradiallypulsatingstar,theradialvelocityhasanodeandthepressurehasanantinode.Attheopenendoftheorganpipeandatthesurfaceofastar,thesituationisreversed:thepressure\ructuationsvanishandthoseofthevelocityaremaximal.(Thelatterboundaryconditionisinaccuratefororganpipes,becauseitdoesn'tallowforradiationofsound,butisbetterforstars.)Inthisapproximation,thelengthofthepipe,a,isonequarterofthewavelengthofthefundamental.Thusthelowestnote|thefundamental|hasperiod0=4a=vs.Thesoundspeedofthestarvarieswithradius,andfurthermoregravitationalforcesmodifytheperiod,butneverthelesswemayexpectthat04R vs:(5)Theequationofhydrostaticequlibrium,dP dr=GMr r2;(6)canbeverycrudelyapproximatedbyP RGM R2;orP GM R;(7)wherePandrepresenttheaveragepressureandtheaveragedensityinthestar.ThecombinationP=isrelatedtothesoundspeed,vs:v2s@P @S1P :(8)Sincetheadiabaticindex1isalwaysoforderunity,itfollowsthatvsGM R1=2:(9)Puttingthisintoeq.(5),we ndthatthefundamentalfrequencyisoforder04GM R31=2;(10)inagreementwitheq.(3).Infact,forann=3=2polytropewith1=5=3,theroughestimate(10)isremarkablyaccurate:theactualnumericalcoecientturnsouttobe3:815insteadof4.Forastarhavingtheequilibriumstructureofann=3polytropebutrespondingtoadiabaticperturbationswith1=5=3(thisisacrudemodeloftheSun),thecoecientis2:065[[1],Table8.1].ByvariationalmethodsakintoRayleigh-Ritz,whichyoumayhaveencounteredinquantummechanics,itispossibletoestablishthefollowingboundsontheangularfrequency(!02=0)ofthefundamentalmode:(314)GM R3!20(314)jWj I;(11)2 Figure1:TheCarnotcycle.inwhichWZGm rdm(12)isthegravitationalpotentialenergyofthestarand2 3I2 3Zr2dm(13)isthemomentofinertia.Foraderivation,seeChapter8of[1].Itisassumedherethatpressureanddensity\ructuationsareconnectedbytheadiabaticrelation(8)with1constantthroughoutthestar.Thelefthandinequalityineq.(11)mustbereversedif14=3.Inthiscasebothlimitsimplythat!20,sothatthefundamentalmodegrowsexponentially.LikeawhitedwarfabovetheChandrasekharmass,suchastarisunstabletogravitationalcollapse.Inanorganpipe,higherharmonicsexistinadditiontothefundamental,withwavelengthsn=4a=(2n+1)andperiodsn=n=vs.Becausethesoundspeedvarieswithradiusinagivenstar,itspulsationalharmonicsarenotsosimplyrelated.However,theperiodsdecreasemonotonicallywithincreasingnumbersofnodes.Thenthovertonehasnnodesinthevelocityordisplacement,notcountingthenodethatisalwayspresentatr=0.Theratio1=0oftheperiodofthe rstovertonetothatofthefundamentalmeasuresthedegreeofcentralconcentrationofthestar;thisratiotendstoincreaseasthecentraldensityincreaseswithrespecttothemeandensity.Forexample,1=0is0:465and0:738forthe(n;1)=(3=2;5=3)and(3;5=3)polytropes,respectively([1]).Excitationofradialpulsations:TheKappaandEpsilonMechanismsIftheadiabaticapproximationwereexact,orinotherwordsiftheentropyofeverygaselementwereconstantthroughthepulsationcycle,thentheenergyinpulsationswouldbeconserved.Oscil-lationsofstarswith1&#x-5.1;䝀4=3throughoutmostoftheirinteriorwouldneithergrownordecay.Onemightthinkthatnonadiabatice ects,suchasviscosityandradiativedi usion,wouldtendtodamptheoscillations.Thisisindeedusuallythecase.Undercertaincircumstances,however,nonadiabatice ectsmayactuallyexcitepulsations.Toseehowthiscanhappen,itisusefultoconvertourorganpipeoftheprevioussectionintoaheatengine.Filltheopenendwithamovablepistonandallowthepipetomakethermalcontactwithhotandcoldheatbaths.Justtoremindyou,theclassicalCarnotengineworksasfollows(refertoFigure1):3 A!B:Thepistoniscompressedadiabatically,thatis,atconstantentropy:SAB=0.PressureP,volumeV,andtemperatureVvaryasP/V1andT/V11.ThetemperatureincreasesfromTAtoTB=TA(VA=VB)11.B!C:Thegasisbroughtintocontactwiththehotterheatbath,Thot=TB,andallowedtoexpandatconstanttemperature.SincePV/TandTisconstant,P/V1.Duringthisstep,theentropyincreasesbySBC=QBC=Thot,whereQBCistheheataddedtothegas.C!D:Thegasisdetachedfromtheheatbathandallowedtoexpandadiabaticallyuntilitstem-peraturefallsbacktoTA=Tcold.AtthispointthevolumeislargerthanVAbecauseoftheincreaseinentropyduringthepreviousstep.D!A:Finally,thegasisbroughtintocontactwiththecoldbathattemperatureTcoldandisothermallyrecompressedtoitsoriginalvolumeVA.Inthisstep,SDA=QDA=Tcold.SinceS(afunctionofstate)mustreturntoitsoriginalvalue,SDA=SBC.SoQBC+QDA=QBC(ThotTcold)=Thot�0.Thusnetheatisabsorbedbytheengine,andenergyconservationrequiresthattheheatisconvertedtowork.ThenetworkdonebythegasonthepistonisW=IABCDAPdV;(14)whichistheareaenclosedbythecurveinthe(V;P)plane.Sincetheinternalenergy(U)returnstoitsoriginalvalueafteronecompletecycle,HdU=0.ThereforeusingtheFirstLaw,dU=TdSPdV,wecanrewritetheworkdoneasW=ITdS:(15)ClearlyW�0alonganyclockwiseclosedpath,notjustthosecomposedofadiabatsandisotherms.Henceifthegastendstogainentropywhenitiscompressed(athightemperature)andtoloseitwhenexpanded(atlowertemperature),thenthenetworkdoneispositive.IntheCarnotenginedescribedabove,heatandentropyaregainedorlostonlywhenthegasisincontactwithaheatbath.Atthecenterofastar,however,heatandentropyarecontinuallyproducedbynuclearreactions.Inequilibrium,heatproductioninthecoreisexactlybalancedbyheatlossfromthesurface,i.e.,bythestellarluminosity.Stellarpulsationdisturbsthisbalanceby(i)modulatingthenuclearreactionrateinthecore(epsilonmechanism);andmoreimportantly,(ii)bymodulatingtheradiativeluminosity(kappamechanism).E ect(i)isalwaysdestabilizingbecausethenuclearreactionrateincreaseswithincreasingtemperatureandpressure:thusthechangeinreactionratetendstoaddentropytothecorewhenitisathigherthanequilibriumtemperature.Exceptinverymassivestars(&100M ),theepsilonmechanismtendstobeweak,sothatasmallamountofdissipationisenoughtopreventgrowth.Atthelowertemperaturescharacteristicoftheouterenvelopesofstars,theopacitymayactuallyincreasewithtemperature(at xedentropy,forexample)becausethegasisincompletelyionized.Insuchacase,luminositydecreaseswhenthestarcontractsanditstemperaturerises,trappingheatintheinteriorandraisingtheentropy;thereverseistruewhenthestarexpands.Duringoneperiodofoscillation,mostmassshellsthenexecuteaclockwiseclosedpathinthe(S;T)and(V;P)planes,leadingtoanincreaseinmechanicalenergy.Thereasonfortheincreaseofopacitywithtemperatureisroughlyasfollows.Inthesolarphotosphere,almostallofthehydrogenatomsareintheirgroundstate(n=1).Toexcitesuchanatomintothehigherstatesrequiresphotonsenergiesof10:2eV,whereasthecharacteristicphotonenergyat5700KisonlykT=0:49eV.Henceneutralhydrogencontributeslittletotheopacity,whichisinfactdominatedbytheweaklyboundHion.Atmodestlyhighertemperatures104K,asmallbutsigni cantproportionofthehydrogenliesinexcitedstates,wherethespacing4 betweenadjacentlevelsissmaller(/n3),asistheionizationenergy(/n2),sothattheopacityisdominatedbytheexcitedneutralhydrogen.Atthesetemperatures,@=@T�0.Atstillhighertemperatures,thefractionofthehydrogeninexcitedboundstatesactuallydecreaseswithincreasingtemperaturebecausemostofthehydrogenisalreadyionized;sincefreeelectronsaremuchlesse ectivethanboundonesatabsorbingphotons,thehydrogenopacitythendecreaseswithfurtherincreasesinT.However,atT4104K,heliumisintransitionbetweensinglyanddoublyionizedstatesandtendstocause@=@T�0.Theseopacitye ectsconstitutethekappamechanism(ii).Wecanunderstandthephysicsinabitmoredetailbyreturningtotheworkintegralintheform(15).ThegaswithintheCarnotengineisuniformintemperature,soSrepresentsthetotalentropyofthisgas.Inthecaseofastar,sincethetemperaturevarieswithradius,wemustevaluateTdSseparatelyforeachmassshell;henceS(m;t)becomestheentropyperunitmassatmassfractionm(thisnotationwillbeusedinplaceofMrinthislecture),andW=t0+Zt0dtMZ0dmTdS dt:(16)ReplaceT(m;t)!T(m)+T(m;t)andS(m;t)!S(m)+S(m;t),whereT(m)andS(m)arethetime-averagedtemperatureandentropyoftheshell,andTandShavezerotimeaverage.ThentheworkintegralbecomesW=IdtZdmTdS dt:(17)Thelimitsofintegrationarethesameasbeforebutwillnotbewrittenoutexplicitly.Noticethatthisissecondorderintheoscillatingquantities;thezeroth-ordertermsvanishbecausedS=dt=0,andthe rst-ordertermsvanishinthetimeaverage:TdS dt=TdS dt=0:Inthefundamentalradialmode,theentirestarexpandsandcontractstogether.HencethesignofTisindependentofmassfraction,andsincethepulsationisapproximatelyhomologous|i.e.,thefractionalchangeinradiusr=randvolumeV=Visapproximatelythesameformostmassshells|eventhemagnitudeofT Tisapproximatelyconstantwithmassfraction.Thusitisusefultopullthefractionaltemperatureperturbationoutofthemassintegral:WIdtT TmZdmTdS dt;(18)wheretheanglebracketsindicateanappropriateaverageovermassattimet.Finally,ifweignorechangesinthenuclearreactionrate,thento rstorder,TdS dt@L @m(m;t):(19)Themassintegralineq.(18)nowyieldsL(r;M)L(r;0),andthis=L(r;M)sinceL(r;0)=0evenunderperturbation.WritingL(t)L(M;t)forthechangeinthetotalstellarluminosity,wehaveatlastWIdtT TmL(t):(20)Theluminosity,opacity,andtemperaturearerelatedbyL(m;t)=162r4c @Prad @m;(21)5 wherePrad=1 3aT4.Atsucientdepthbelowthephotosphere,wemayevaluateLfromthisequationusingtheadiabaticvaluesforT,,andr.Variationsinopacity,radius,andradiativegradientcontributeadditivelyto rstorder.Clearlyif@ @TS=@ @T+@ @T@ @TS�0;thentheopacitywillincreaseduringcontractionsThesecondtermontherighthandsidetendstobepositivesince(@=@)T0and(@=@T)S�0.The rstcanbepositiveatlowtemperatures:seeFigure17.5inthetextbook[3].ThiswilltendtomaketheproductLT0ineq.(20).Thedecreaseinradiusduringcontractionsreinforcesthistendencybecauseofther4factorineq.(21).Ontheotherhand,Prad&#x-5.1;䘘0incontractionsandthisimplies@Prad=@m0inahomologouscontraction:@ @mPrad=Prad Prad@Prad @m:Sothe rsttwoe ectsmustbestrongerthanthelastinordertoguaranteethatW�0,i.e.,inordertoincreasethemechanicalenergyofthepulsation.Intheargumentabove,weevaluatedthechangeinluminosityusingtheadiabaticvariationsinT,r,and.Thisisonlyanapproximation,sinceitispreciselythedeviationfromadiabaticbehaviorthatmakesgrowthpossible.Theapproximationisvalidifthelocalthermaltimescaleoftheionizationzone(i.e.theregionwhere@=@T�0)islongcomparedtothepulsationperiod.Thelocalthermaltimescalemaybede nedasthethermalenergyofthematterexteriortothebaseoftheionizationzone,dividedbytheluminosity:ttherm(Mm)P L m=mioniz::(22)Ifttherm,thentheadiabaticapproximationfailscompletelyintheionizationzone:theradiativegradientwillquicklycompensateforanychangeinopacitysothattheluminosityisnota ectedbythepulsationandW0.AsTe increases,theionizationzonesmoveclosertothesurfaceandtthermdecreases,andatsomepointtheinstabilityisquenched.Thisdeterminestheblue,i.e.high-Te ,edgeofthe\instabilitystrip"intheHRdiagramwhereCepheidsoccur.Ontheotherhand,iftheionizationzoneliestoofarbelowthesurface,thentthermbothintheionizationzoneandintheregionsimmmediatelyaboveit;modulationofthelocalluminosityintheionizationzonethenhaslittlee ectonL(t)emergingfromthesurface,whichisdeterminedbylocalconditionsinsomeshallowerzonewherettherm.Furthermore,asTe decreasesandtheionizationzonesretreattogreaterdepths,theybecomeconvective,atwhichpointtheopacitynolongercontrolstheradiative\rux.Thesee ectsdeterminetherededgeoftheinstabilitystrip.Ourderivationofthekappamechanismisimpressionisticratherthanrigorous.Theproperwaytodeterminepulsationalstabilityistolinearizethefullnon-steadystellarstructureequationsandsolveforthecomplexeigenfrequenciesdirectly.Thisisdescribedin[1]andwillnotberepeatedhere.DirectDeterminationofDistance:TheBaade-WesselinkMethodWehavenotedthatonceaperiod-luminosityrelationhasbeenestablished,pulsatingstarscanserveasdistanceindicators.Remarkably,however,thedistancetoapulsatingstarcanbedetermineddirectly,inprinciple,withoutpriorknowledgeofthePLrelation.The\ruxreceivedfromastarofradiusR,distanceD,ande ectivetemperatureTe isF=R2T4e D2:(23)The\ruxande ectivetemperatureofapulsatingstarvarywithtime,butthesearedirectlymea-surableifwehavephotometryinmultiple lters,orbetteryetseriesofspectra.Takingthelogofeq.(23)anddi erentiating,onehasd dtlnF=4d dtlnTe +2 RdR dt:(24)6 Figure2:LightcurvesofafewoftheLMCCepheidsfoundbytheOGLEIIproject[7].Noticethatthedistancehasdroppedoutbecauseitisconstant.Onecansolvetheaboveforthestellarradiusintermsofmeasurablequantities:R=2dR dtd dtlnF4d dtlnTe 1:(25)HeredR=dtismeasuredbythedopplershiftofstellarabsorptionoremissionlines.OnceRisknown,eq.(23)canbesolvedforD.Anobviousconsistencycheckisthatthederiveddistanceshouldnotvaryoverthepulsationcycle.Thiselegantmethodsu erssomepracticaldiculties.First,determiningTe fromcolorsorspectraisnottrivial.Thestardoesnotradiateasablackbody|ifitdid,therewouldbenospectrallineswithwhichtomeasuretheradialvelocity.Second,thespectrallinesusedforthevelocityandthecontinuumrepresentedbyTe arenotformedatquitethesameradius.Thismeansofcoursethatthelinesandthecontinuummayhavedi erentvelocities.Bothdicultiescanbeconqueredinprinciplebydetailedphysicalmodelingofthemovingstellaratmosphere,butthisiscomputationallychallenging.The eldisstillunderdevelopment.Incidentally,eq.(25)doesnotrequirethatthepulsationbeperiodic.ThesametechniquehasbeenusedtomeasuredistancestoextragalacticTypeIIsupernovae,butthisapplicationiscalledbyitspractitioners\TheExpandingPhotosphereMethod"[[6],[2]].7 Figure3:I-bandperiod-luminosityrelationofthefundamental-modeCepheidsintheLMC(upperpoints)andSMC(pointsshifteddownby0:8mag)[8].CalculationoflinearadiabaticradialpulsationsTodeterminethepulsationaleigenmodesofagivenstellarmodel,onestartswiththefullradialforceequationintheform@2r @t2m=4r2@P @m+Gm 4r4:(26)Thereisanequilibriumstateinwhichr(m;t)=r0(m)andbothsidesoftheequationvanish.De ner(m;t)r(m;t)r0(m),andsimilarlyP(m;t),(m;t),etc.Presumingthattheseperturbationsaresmall,expand(26)to rstorder,obtainingr=4r2@P @m4Gm 4r5r:Thefollowingequationsareslightlytidierifr0ratherthanmisusedasindependentvariable.Tosavewriting,thesubscriptwillbeomittedfromr0,butitshouldberememberedthatunlesspre xedby,rwillrefertotheequilibriumradiusofagivenmassshell.Thenthelastequationbecomesr=1 @P @r+4GMr r3r:(27)Wehaveputm!Mrtoremindourselvesthatitreferstothemasscontainedwithintheequilibriumradiusr.Thedotsrefertoderivativeswithrespecttotatconstantmorr0.Asyet,wehavenotusedtheassumptionofadiabaticity,butweareabouttodoso:P=@P @S=1P ;where1@lnP @lnS:(28)Inagas-pressure-dominatedstar,15=3everywhere|inradiativeaswellasconvectiveregions|because1describesnottheequilibrium(inwhichentropymayvarywithradius)buttheresponseofagivenmasselementtoadiabaticperturbations.Conservationofmassimplies =1 r2@ @rr2r:(29)8 Bycombining(28)and(29),oneconverts(27)toanequationinvolvingralone:r=1 @ @r1P r2@ @rr2r+4GMr r3r:(30)Foraneigenmode,allmassshellsshouldvibrateatthesamefrequency:thatis,r(m;t)=(r)ei!t;sothat!2=1 d dr1P r2d drr2+4GMr r3:(31)Fornumericalwork,itisoftenconvenienttounpack(31)asapairofordinarydi erentialequationsbyintroducingr2d(r2)=dr,whichisessentiallyjust=,sothatd dr=2 r;1 d dr(1P)=!24GMr r3:(32)Theboundaryconditionatr=0isobviously=r=0.Theboundaryconditionatr=R(theequilibriumouterradius)isnotsoobvious,butitcanbeobtainedfromtherequirementthatshouldberegularasr!R:thatis,(r)andatleastits rstfewderivativeswithrespecttorshouldexistatr=R.[Infact,therequirementofregularityasr!0givesthecentralb.c.,(0)=0,ascanbeseenfromthe rstofequations(32).]Letusrewritethesecondofequations(32)intheform1P d dr1GMr r2=!2+4GMr r3:(33)Thecoecientofd=drinthisequationisv2s,whichisproportionaltotemperatureforanidealgasandthereforeverysmallatthesurfacecomparedtoitsvalueatthecenter.Infact,forthepurposeofcomputingmodesofoscillation,oneoftenapproximatesthehydrostaticequilibriumstructurebymodelsinwhichv2s!0asr!R.Ifthesolutionistoberegular,thend=drwillbe niteasr!R;infactd=dr=R2there.Sothe rsttermonthelefthandsideof(33)essentiallyvanishesatr=R,andtheboundaryconditionsbecome=1 1RR3!2 GM+4atr=R(34)=0atr=0:(35)Thesurfaceboundarycondition(34)isonlyapproximateandwouldbreakdownforhigh-frequencymodeswhosewavelengthisshorterthanthepressurescaleheightatthephotosphere;intheSun,thisoccursfor.3min,whereas0; 1hr.Wemaysolveequations(32),(34)&(35)byshootingtoa ttingpoint,i.e.integratingoutwardfromr=0andinwardfromr=Rtoanintermediateradiusr tButwehavethreeunknownstoguess:(0),(R),and!itself.Fortunately,becausethedi erentialequationsarelinearandhomogeneous,wedonothavetosolveforallthreesimultaneously.Wecanfreelyrescalethedependentvariablesbyacommonfactor|andwemaychoosethesescalingfactorsindependentlyfortheinwardandoutwardintegrationstominimizethemismatchatr t.Thisfreedomreducesthenumberofmatchingconditionsfromtwotoone: outward= inwardatr=r t:(36)Sincebothsidesofthisequationareindependentofthescalingfactors,wemaychoose(0)and(R)arbitrarily|Igenerallyjustsetbothtounity.Thenwehaveaone-dimensionalsearchover!2toperforminordertosatisfy(36).Thelowestrootfor!2isthefundamentalmode.After nding!,wemayeasilyrescale(0)and(R)soastoobtaincontinuoussolutionsforand.9 References[1]J.P.Cox.TheoryofStellarPulsation.PrincetonUniversityPress,1980.[2]R.G.Eastman,B.P.Schmidt,andR.Kirshner.TheAtmospheresofTypeIISupernovaeandtheExpandingPhotosphereMethod.ApJ,466:911{937,August1996.[3]R.KippenhahnandA.Weigert.StellarStructureandEvolution.Springer,1990.[4]H.Leavitt.1777VariablesintheMagellanicClouds.Ann.HarvardColl.Obs.,60(4),1908.[5]B.F.MadoreandW.L.Freedman.TheCepheiddistancescale.PASP,103:933{957,September1991.[6]B.P.Schmidt,R.P.Kirshner,andR.G.Eastman.ExpandingphotospheresoftypeIIsupernovaeandtheextragalacticdistancescale.ApJ,395:366{386,August1992.[7]A.Udalski,I.Soszynski,M.Szymanski,M.Kubiak,G.Pietrzynski,P.Wozniak,andK.Zebrun.TheOpticalGravitationalLensingExperiment.CepheidsintheMagellanicClouds.IV.CatalogofCepheidsfromtheLargeMagellanicCloud.ActaAstronomica,49:223{317,September1999.[8]A.Udalski,M.Szymanski,M.Kubiak,G.Pietrzynski,I.Soszynski,P.Wozniak,andK.Zebrun.TheOpticalGravitationalLensingExperiment.CepheidsintheMagellanicClouds.III.Period-Luminosity-ColorandPeriod-LuminosityRelationsofClassicalCepheids.ActaAstronomica,49:201{221,September1999.10

Related Contents


Next Show more