ebook img

Evolution of massive stars at very low metallicity including rotation and binary interactions PDF

0.19 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Evolution of massive stars at very low metallicity including rotation and binary interactions

Evolution of massive stars at very low metallicity including rotation and binary interactions S.-C Yoon∗, M. Cantiello† and N. Langer† ∗DepartmentofAstronomyandAstrophysics,UniversityofCalifornia,SantaCruz,CA95064,USA †AstronomicalInstitute,UtrechtUniversity,Princetonplein5,3584CC,Utrecht,TheNetherlands 8 0 0 Abstract. Wediscussrecentmodelsontheevolutionofmassivestarsatverylowmetallicityincludingtheeffectsofrotation, 2 magnetic fieldsandbinarity. Verymetal poor starsloseverylittlemassand angular momentum duringthemainsequence evolution, and rotation plays a dominant role in their evolution. In rapidly rotating massive stars, the rotationally induced n mixing time scale can be even shorter than the nuclear time scale throughout the main sequence. The consequent quasi- a chemically homogeneous evolution greatly differs from the standard massive star evolution that leads to formation of red J giants with strong chemical stratification. Interesting outcomes of such a new mode of evolution include the formation of 8 rapidlyrotatingmassiveWolf-Rayetstarsthatemitlargeamountsof ionizingphotons, theformationofalonggamma-ray 2 bursts and a hypernovae, and the production of large amounts of primary nitrogen. We show that binary interactions can furtherenhancetheeffectsofrotation,asmassaccretioninaclosebinaryspinsupthesecondary. ] h Keywords: Stars:evolution,Stars:rotation,Stars:binary p PACS: 43.35.Ei,78.60.Mq - o r 1. INTRODUCTION large amount of angular momentum at lower metallic- t s ity,resultingintheabundantproductionofenergeticsu- a Massive stars affect the evolution of the early universe pernovaeand gamma-raybursts (Yoon& Langer 2005; [ in a number of ways. They are believed to be the main Woosley&Heger2006;Yoon,Langer&Norman2006). 1 source of reionizing photons at high redshift, and their Detailed numerical simulations of the evolution of v explosionsprovidelargeamountsofenergyintothesur- massive stars including the effects of rotation involve 3 rounding medium. Most elements heavier than helium manyuncertainphysicalprocessessuchasthetransport 7 begintoexistintheearlyuniverseduetonucleosynthe- of angular momentum and chemical species due to ro- 3 sis in massive stars. The history of star formation and tationally induced instabilities. Studies by Langer et al. 4 . theevolutionofgalaxiesintheearlyuniversearethere- (1999), Heger, Langer & Woosley (2000) and Hirschi, 1 forecloselyrelatedtosuchfeedbackfrommassivestars. Meynet & Maeder (2004) indicate that their adopted 0 Thishasmotivatedmanytheoreticalstudiesontheevolu- angular momentum transport mechanisms (Eddington 8 0 tionofmassivestarsofthefirstandsecondgenerations, Sweet circulations, shear instability and baroclinic in- : whicharecharacterizedbyzero/lowmetallicity. stability)aretooinefficienttoexplaintheobservedspin v Mass loss due to stellar winds from massive stars rates of white dwarfs and young neutron stars: their i X is mainly driven by metal lines (Castor, Abbott & models predict one or two orders of magnitude higher r Klein 1975; Pauldrach, Puls & Kudritzki R.P. 1986; spin rates than the observed values. More recent mod- a Vink,deKoter&Lamers2001).Asaresult,metalpoor els adopting the prescription by Spruit (2000) for the massive stars are expected to lose only a small amount Tayler-Spruitdynamoin differentiallyrotatingradiative of massand angularmomentumandto be keptrotating layers are shown to be more consistent with observa- rapidly,duringtheirlifetimes(seeEkströmandMeynet, tionsintermsofthespinratesofstellarremnants(Heger, however,inthisvolume).Rotation–oneoftheprimary Woosley&Spruit2005;Suijsetal.inprep.).Thevalid- factors to determine the evolution of massive stars – ity of the Talyer-Spruitdynamohas recentlybeen chal- may thus play an even more important role in the evo- lenged by several authors (Denissenkov, Pavel, & Pin- lution of massive stars in the early universe, compared sonneault 2007; Zahn, Brun & Mathis, S. 2007) but it to the case of galactic metal-rich massive stars. For in- seems clear that an efficient braking mechanism that is stance, the high ratio of nitrogen to carbon abundance comparabletowhatSpruitsuggestsisneededtoexplain observed in extremely metal poor stars may be related observations. tochemicalmixingdueto rotationallyinducedinstabil- Here we present recent models of both single and ities in massive stars in the early universe(Chiappiniet binary massive stars at very low metallicity including al. 2006). Stellar cores could also more easily retain a the Tayler-Spruit dynamo as well as other effects of chemical mixing by Eddington Sweet circulations and the classical core-envelope structure is not developed (Maeder1987;Langer1992).Suchastarwouldbegrad- uallytransformedintoarapidlyrotatingmassivehelium star(i.e.,WRstar)bytheendofthecorehydrogenburn- ing.Thismodeofevolutionissupposedtobeparticularly important for metal poor massive stars. At high metal- licity, strong winds significantly spin down a massive star early on the main sequence, rendering rotationally inducedmixingtooinefficientto lead to the chemically homogeneousevolution(seeFig.1). In Fig. 2, we summarize how the evolution and fi- nal fate of massive stars differ according to the initial mass and rotational velocity, at two different metallici- FIGURE 1. Total angular momentum in 20 M⊙ models as ties,basedonourstellarevolutionmodels(Yoon,Langer afunctionofevolutionarytime.Thesolidlineandthedashed & Norman2006).Slowly rotatingstars follow the clas- linesgivetheresultswithZ=10−5and0.02respectively. sical evolutionary path such that they evolve redwards on the HR diagram, and end their life as a (super) gi- antwithathickhydrogenenvelope.Inthiscase,thecore rotation such as rotationally induced chemical mixing, is strongly braked down during the giant phase by the enhancedmasslossnearthebreak-upvelocityandtidal heavy envelope,resulting in a specific angular momen- interactions.We suggestthatrotationcouldleadtovery tum of the order of 1014 cm2s−1 in the pre-collapsing different types of massive star evolution at very low core.TypeIIsupernovaearetheexpectedoutcome,leav- metallicity, compared to the case of metal rich stars. ing millisecond pulsars or black holes as remnants de- Implicationsforsupernovaeand gamma-raybursts, and pendingontheinitialmass. massive star feedback in the early universe are briefly Ontheotherhand,thequasi-chemicallyhomogeneous discussed. evolutionofrapidlyrotatingstarshaveseveralinteresting consequences. Avoiding the giant phase, some of them can retain a large amount of angular momentum in the 2. EVOLUTIONOFSINGLE STARS AT core (j ≥∼1016 cm2s−1) that may lead to produc- CO VERY LOWMETALLICITY tion of a long gamma-rayburst (GRB) or a jet-induced energetic supernova/hypernova (Yoon & Langer 2005; It is well knownthat manymassive stars in our Galaxy Woosley&Heger2006).AhigherratioofGRBs/HNeto as well as in the Small/Large Magellanic Cloud are SNe is predictedfor lowermetallicity (Yoon,Langer& rapid rotators (e.g. Maeder & Meynet 2000; Mokiem Norman2006),andtheroleoftheGRBs/HNefeedback et al. 2006; Mokiem et al. 2007). One of the most im- maybesignificantintheearlyuniverse.E.g.,suchener- portant effects of rotation on the evolution of massive geticsupernovaemighthaveleftuniquenucleosynthetic stars is rotationally induced mixing of chemical ele- signaturesintheearlyuniverseasdiscussedbyNomoto ments(Maeder&Meynet2000;Heger&Langer2000). (inthisvolume;Tominaga,Umeda&Nomoto2007),and In magnetic models, stars tend to rotate rigidly due couldhaveaffectedthehistoryofstarformationathigh to magnetic torques, and thus it is Eddington-Sweet redshift(Kobayashi,Spiringel&White2007).Thevery circulations that mainly induce chemical mixing in- efficientrotationallyinducedchemicalmixingalsoleads side stars (Maeder & Meynet 2005; Heger, Woosley to abundant production of primary nitrogen as shown & Spruit 2005; Yoon & Langer 2005), rather than the in Fig. 3. It is also remarkable that massive WR stars shearinstabilitythatisimportantinnon-magneticmod- couldbeproducedevenatverylow/zerometallicitydue els(Maeder&Meynet2000;Heger,Langer&Woosley tothequasi-chemicallyhomogeneousevolution,without 2000;Meynet&Maeder2002).Efficientchemicalmix- theneedofstellarwinds.Suchmassivehelumstarspro- ingacrossthe boundarybetweenthecoreandtheenve- duce a few times more hydrogen ionizing photons and lopeinastarbyEddington-Sweetcirculationsisusually about100times morehelium ionizingphotonsthan the prohibited due to the chemical gradient built up by nu- correspondingslowly rotatingstars(Fig. 4). Thismight clear burning in the convective core (Zahn 1992). If a have consequences in the history of reionization in the star is born with a sufficiently large amount of angular earlyuniverse,dependingontheinitialrotationalveloc- momentum, however, mixing could occur on a shorter ityfunctionofthefirstandsecondgenerationsofstars. time scale than the nuclear burning time scale. I.e., the star becomes chemically homogeneous due to efficient FIGURE2. Finalfateofourrotatingmassivestarmodelsattwodifferentmetallicities(Z=0.001&0.00001),intheplaneof initialmassandinitialfractionoftheKeplerianvalueoftheequatorial rotationalvelocity.TobothsidesoftheGRBproduction regionforZ=0.001,blackholesareexpectedtoforminsideWRstars,butthecorespinisinsufficienttoallowGRBproduction. ForZ=0.00001,thepair-instabilitymightoccurtotherightsideoftheGRBproductionregion(seeHegeretal2003),althoughthe rapidrotationmayshiftthepairinstabilityregiontolargermasses.Thedashedlineintheregionofnon-homogeneous evolution separates TypeII supernovae (SN II;left) and blackhole (BH;right) formation, wherethe minimummass for BHformation is simplyassumedtobe30M⊙.FromYoon,Langer&Norman(2006). Langer1999;Wellstein, Langer& Braun2001).Binary starevolutioniscloselyrelatedtothechangeintheradii of the stellar components throughout the evolutionary stages.Atsolarmetallicity,theenvelopeofamassivestar drastically expands during the helium core contraction, and the case B mass transfer is most likely. As metal- licity decreases, the rapid expansion of the envelope of amassivestarisdelayedtoalaterevolutionarystageas revealed in Fig. 5, and the Case C mass transfer would become more common (see de Mink et al. in this vol- ume). In the stellar models at very low metallicity (Z ≤ 10−5),massivestarsremainfairlycompactwiththemax- imumradiuslessthanafewhundredssolarradiithrough- outtheevolutionifovershootingisignored.Thisimplies thata veryclose orbitmaybe requiredfora verymetal poor star to undergo Roche-lobe overflows in a binary system.Ourmodelsalsoindicatethat,atverylowmetal- licity,masstransferisusuallydynamicallystableregard- less of the evolutionarystages of the primary,since the FIGURE 3. CNO elements yields by stellar winds from 3 enveloperemainsradiative.However,rapidexpansionof 2(dfi0iflflaeenrddecnZitrec=vleo1sl)0u,−tMi5oinn(aoitrp=yens4e0sqquauenandrceZess)=.w1it0h−M5(inaistt=eri4s0ksa)n,danZd=Mi1n0it−=8 tthioeneinsvoebloseprevuepdtion∼the10m3oRd⊙eldsuwriinthgZth=eC1O0−c5owrehceonnotvraecr-- shooting is considered (de Mink et al. in this volume), and it remains uncertain whether very metal poor stars 3. EVOLUTION OFBINARYSTARS AT wouldbecomesuper-giants(seealsoWoosleyinthisvol- ume). VERY LOWMETALLICITY A high mass transfer rate in a massive close binary usuallyleadstothermalexpansionoftheenvelopeofthe Massivestarevolutionbecomesmuchcomplicatedbybi- mass-accreting star. This effect becomes, however, less naryinteractions(e.g.de Greve& deLoore1992;Pod- significantforlower metallicity andbinarysystems can siadlowski, Joss & Hsu 1992; Pols 1995; Wellstein & avoidcontactorcommonenvelopephasemoreeasily(de FIGURE4. Totalnumberofhydrogen(leftpanel)andhelium(rightpanel)ionizingphotonsemittedthroughouttheevolutionof massivestarmodelsatZ=10−5,asafunctionoftheinitialmass.Thelinesconnectingasterisksandfilledtrianglesgivetheresults from rapidly rotating stellar models that undergo the quasi-chemically homogeneous evolution (see the text), and non rotating models,respectively. mass accretion should experiencestrong chemicalmix- ing during the rest of its life, unless the orbit remains close enough for the star to be spun down by tidal in- teraction on a short time scale, after the mass transfer phase. If the secondary is rejuvenated sufficiently, even thequasi-chemicallyhomogeneousevolutioncanbe in- duced by mass accretion, eventually leading to produc- tion of a long GRB (Cantiello et al. 2007). This binary star scenario for long GRB progenitors suggests that a significant fraction of long GRBs should occur in run- away stars (see Fig. 6 and Cantiello et al. 2007) as im- plied by a recent observational study of Hammer et al (2006). In short, a large fraction of massive binaries at very lowmetallicitymayundergodynamicallystableCaseC FIGURE 5. Change in the radii of 20 M⊙ models with Vrot,init=0.3VKepleratfourdifferentmetallicities,asafunction masstransfer.Therefore,mostprimariesinclosebinaries ofcentraltemperature. would explode as type Ibc supernovae,while most sec- ondariesastypeIIsupernovae,butsomeofthemaslong GRBs/hypernovae.Theeffectsofchemicalmixingsuch Minketal.,inthisvolume).Therefore,itisexpectedthat as productionof primarynitrogenmay be significantin formationofcompactbinarysystemssuchasblackhole massaccretingsecondarystars. X-ray binaries via common envelope phases should be lesscommonatlowermetallicity. The role of rotation could be enhanced by binary 4. CONCLUDINGREMARKS interaction. A slowly rotating star could be spun up to the break-up velocity by mass accretion (Langer et We still do not fully understand the details about the al.2003;Petrovic,Langer&vanderHucht2005).Mod- effects of rotation such as transport processes of angu- els with fast semi-convection show that rapid mass ac- lar momentum and chemical species due to magnetic cretion leads to rejuvenation in mass-accreting stars, torques,EddingtonSweetcirculationsandotherpossible which weakens the chemical gradient between the core rotationallyinducedhydrodynamicinstabilities in stars, and the envelope(Braun & Langer 1995). All these ef- which are important ingredients in recent stellar evolu- fects of mass accretion, i.e., increase in mass and an- tionmodels.Improvedtreatmentsoftherotation-related gularmomentumandrejuvenation,favorchemicalmix- physicsarethusneededforfuturestudies,inconnection ing by Eddington Sweet circulations in mass-accreting with observational tests and multi-dimensional simula- stars.Therefore,sucharapidlyrotatingstarproducedby tions(e.g.Brottetal.inthisvolume;Talon2007).How- 4. Castor,J.I.,Abbott,D.C.,&Klein,R.I.,1975,ApJ,195, 157 5. Chiappini, C., Hirschi, R., Meynet, G., Ekström, S., Primary Secondary Maeder,A.,&Matteucci,F.,2006,A&A,449,27 6. Denissenkov,P.A.,&Pinsonneault,M.,2007,ApJ,655, 1157 7. deGreve,J.P.,&deLoore,C.,1992,A&AS,96,653 8. Hammer,F.,Flores,H.,Schaerer,D.,Dessauges-Zavadsky, M.,LeFloch,E.,&Puech,M.,2006,A&A,454,103 9. Heger,A.,&Langer,N.,2000,ApJ,544,1016 10. Heger,A.,Langer,N.,&Woosley,S.E.,2000,ApJ,528, 368 11. Heger,A.,Woosley,S.E.,&Spruit,H.C.,2005,ApJ,626, 350 WR star 12. Hirschi,R.,Meynet,G.,&Maeder,A.,2004,A&A,425, Secondary 649 13. Kobayashi, C., Springel,V., &White,S.D.M.,2007, MNRAS,376,1465 14. Langer,N.,1992,A&A,265,17 16. Langer,N.,Heger,A.,Wellstein,S.,&Herwig,F.,1999, A&A,346,37 16. Langer,N.,Yoon,S.-C.,Petrovic,J.,&Heger,A.,2003,in StellarRotation,Proc.IAU-Symp.215,ASP,SanFrancisco, SN explosion A.Maeder,P.Eenens,eds. (cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:0)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1)(cid:1) Secondary Run away 111789... MMMaaaeeedddeeerrr,,,AAA...,,,1&&98MM7ee,yyAnn&eettA,,GG,1..,,722800,001055,,9AAR&AA&,4A4,03,81,014413 20. Meynet,G.,&Maeder,A.,2003,A&A,390,561 SN Kick 21. Mokiem,M.R.,deKoter,A.,&Evans,C.J.etal.,2006, A&A,456,1131 FIGURE 6. Runaway star scenario for long GRB progeni- 22. Mokiem,M.R.,deKoter,A.,&Evans,C.J.etal.,2007, tors.Inaclosemetalpoorbinarysystem,thesecondarygains A&A,465,1003 mass and angular momentum from the primary. If the sec- 23. Spruit,H.C.,2002,A&A,381,923 ondary is rejuvenated and spun up sufficiently as a result, it 24. Pauldrach, A.W.A.,Puls,J.,&Kudritzki, R.P.,1986, undergoes the quasi-chemically homogeneous evolution. The A&A,164,86 primarybecomesaWRstarandexplodeasatypeIb/csuper- 25. Petrovic,J.,Langer,N.,&vanderHucht,K.A.,2005, nova.Ifthesupernovakickunboundsthesystem,thesecondary A&A,435,1013 runsawayatavelocityof∼10...100km/s,andexplodeasa 26. Petrivic,J.,Langer,N.,Yoon,S.-C.,&Heger,A.,2005, longGRB/hypernovaaftertravelingafewtoseveralhundreds A&A,435,247 PCsaway.SeeCantielloetal.(2007)fordetaileddiscussions. 27. Podsiadlowski,Ph.,Joss,P.C.,&Hsu,J.J.L.,1992,ApJ, 391,246 28. Pols,O.R.,1994,A&A,290,119 ever,abovediscussionsclearlyindicatethattheevolution 29. Talon,S,2007,[astro-ph/0708.1499] of metal poor massive stars much depends on rotation. 30. Tominaga,N.,Umeda,H.,&Nomoto,K.,2007,ApJ,660, 516 Addressingtheroleofthemassivestar feedbackin the 31. Vink,J.S.,deKoter,A.,&Lamers,H.J.G.L.M.,2001, earlyuniverseincludingtheeffectsofrotationandbina- A&A,369,574 rity will be an exciting but challenging subject for the 32. Wellstein,S.,&Langer,N.,1999,A&A,350,148 next decade. In particular, future observational studies 33. Wellstein,S.,Langer,N.,&Braun,H.,2001,A&A,369, on massive star populations (e.g. WR/O ratio) in metal 939 poorgalaxiesandsupernovae/GRBsathighredshiftmay 34. Woosley,S.E.,&Heger,A.,2006,ApJ,637,914 35. Yoon,S.-C.,&Langer,N.,2005,A&A,443,643 providestrongconstraintsfortheoreticalmodels,asdis- 36. Yoon,S.-C.,Langer,N.,&Norman,C.,2006,A&A,460, cussedinYoon,Langer&Norman(2006)ingreaterde- 199 tail. 37. Zahn,J.-P.,1992,A&A,265,115 38. Zahn,J.-P.,Brun,A.S.,&Mathis,S.,2007, A&A,in press,[astro-ph/0707.3287] REFERENCES 1. Braithwaite,J.,2006,A&A,449,451 2. Braun,H.,&Langer,N.,1995,A&A,297,483 3. Cantiello,M.,Yoon,S.-C.,Langer,N.&Livio,M.,2007, A&A,465,L29-L33

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.