Table Of Content

Astronomy&Astrophysicsmanuscriptno.arxiv (cid:13)c ESO2010 January10,2010 Predicted gamma-ray line emission from the Cygnus complex PierrickMartin1,2,Ju¨rgenKno¨dlseder1,2,GeorgesMeynet3,andRolandDiehl4 1 Centred’EtudeSpatialedesRayonnements,CNRS/UPS,9,avenuecolonelRoche,BP44346,31028Toulousecedex4,France 2 Universite´deToulouse(UPS),CentreNationaldelaRechercheScientifique(CNRS),UMR5187,France 3 GenevaObservatory,CH-1290Sauverny,Switzerland 4 MaxPlanckInstitutfu¨rextraterrestrischePhysik,Postfach1312,85741Garching,Germany Received:10July2009/Accepted:17November2009 0 ABSTRACT 1 0 Context.TheCygnusregionharboursahugecomplexofmassivestarsatadistanceof1.0-2.0kpcfromus.About170Ostarsare 2 distributedoverseveralOBassociations,amongwhichtheCygOB2clusterisbyfarthemostimportantwithabout100-120Ostars. n Thesemassivestarsinjectlargequantitiesofradioactivenucleiintotheinterstellarmedium,suchas26Aland60Fe,andtheirgamma- a raylinedecaysignalscanprovideinsightintothephysicsofmassivestarsandcore-collapsesupernovae. J Aims.PaststudiesofthenucleosynthesisactivityofCygnushaveconcludedthatthelevelof26Aldecayemissionasdeducedfrom CGRO/COMPTELobservationswasafactor2-3abovethepredictionsbasedonthetheoreticalyieldsavailableatthattimeandon 0 theobservedstellarcontentoftheCygnusregion.Wereevaluatethesituationfromnewmeasurementsofthegamma-raydecayfluxes 1 withINTEGRAL/SPI(presentedinapreviouspaper)andnewpredictionsbasedonrecentlyimprovedstellarmodels. Methods.Webuiltagridofnucleosynthesisyieldsfromrecentmodelsofmassivestars.Comparedtopreviousworks,ourdatain- ] E cludesomeoftheeffectsofstellarrotationfor thehigher massstarsandacoherent estimateofthecontributionfromSNIb/c.We H thendevelopedapopulationsynthesiscodetopredictthenucleosynthesisactivityandcorrespondingdecayfluxesofagivenstellar populationofmassivestars. . h Results.The observed decay fluxesfrom theCygnus complex arefound to beconsistent withthe values predicted by population p synthesisatsolarmetallicity;andyet,whenextrapolatedtothepossiblesubsolarmetallicityoftheCygnuscomplex,ourpredictions - failtoaccount fortheINTEGRAL/SPImeasurements. Theobservedextent ofthe1809keV emissionfromCygnus isfoundtobe o consistentwiththeresultofanumericalsimulationofthediffusionof26AlinsidethesuperbubbleblownbyCygOB2. r Conclusions.Ourworkindicatesthatthepastdilemmaregardingthegamma-raylineemissionfromCygnusresultedfromanover- t s estimateof the1809keV flux of the Cygnus complex, combined withanunderestimate of the nucleosynthesis yields. Our results a illustratetheimportanceofstellarrotationandSNIb/cinthenucleosynthesisof26Aland60Fe.Theeffectsofbinarityandmetallicity [ mayalsobenecessarytoaccountfortheobservationssatisfactorily. 1 Key words. Gamma rays: observations – open clustersand associations: individual: Cyg OB – Stars:early-type – ISM:bubbles – v Nuclearreactions,nucleosynthesis,abundances 2 2 5 1. Introduction thereforeinevitableinastrophysics.Amongstthevariousobser- 1 . vationalapproachesemployedtoexplorethephysicsofmassive 1 stars and their SNe, gamma-ray line astronomy can be a very 0 Conventionally defined as stars with initial masses above 8-10 valuabletool. By studyingthe characteristic decay radiationof 0 and upto about120M , massive stars play a fundamentalrole ⊙ certainradio-nuclidesproducedinthehydrostaticand/orexplo- 1 inastrophysicsbecauseoftheverypowerfulbehaviourtheyex- siveburningstagesofmassivestars,gamma-raylineastronomy : hibit during their lives, from the early main-sequence to their v providesindirectaccesstothephysicalconditionsthatgoverned Xi finalexplosion(andevenbeyondthankstothecompactobjects their synthesis and ejection in the ISM. Among the few radio- they leave behind). Usually clustered in OB associations of a isotopesaccessibletothepresent-daygamma-raytelescopes,the r few tens to several thousand members, massive stars severely a long-lived26Aland60Feholdpotentialforinterestingconstraints affect their environment on the hundred-parsec scale through ontheevolutionofmassivestars. their sustained mechanical and radiative energy output: strong TheGalacticdecayemissionfrom26Alwasextensivelystud- luminosity in the Lyman continuum, very energetic winds and iedinthe1990swiththeCGRO/COMPTELinstrumentandthe masslossepisodes,andsupernovaexplosions.Inadditiontome- resultant cartography of the 26Al decay signal at 1809keV re- chanicalenergy,thesestellaroutflowsalsocarrytheproductsof vealed that most of the emission comes from the inner Galaxy nuclear burning and thus enrich the interstellar medium (ISM) and from a few nearby star-forming regions(Diehletal. 1994; with heavy elements. The physical and chemical evolution of delRioetal. 1996; Plu¨schkeetal. 2001). The intensity distri- the Galaxy is driven to a considerable extent by massive stars bution observed with COMPTEL allowed it to be established and a wide varietyof Galactic objectsfollow fromtheir evolu- thatmassivestarsarethedominantsourceof26AlintheGalaxy tion:supernovae(SNe)andtheirremnants(SNRs),neutronstars (Prantzos&Diehl 1996; Kno¨dlsederetal. 1999). In the early orstellar-massblackholes,someofwhichmanifestintheform 2000s, the first detection of the Galactic 60Fe decay signal at ofpulsarsormagnetars,X-raybinaries,ormicroquasars. 1173 and 1332keV was achieved from RHESSI observations Athoroughunderstandingoftheprocessesthatruletheevo- (Smith2004),withameasuredfluxofabout15%ofthe26Alde- lution of massive stars and their death as core-collapse SNe is 2 PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex cayflux.Althoughthedecayemissionwastooweaktoachieve ies have been devoted to the nucleosynthesis of 26Al and a detailed cartography and thus confirm that massive stars are 60Fe in the Cygnus region (see for example delRioetal. themainsourceof60Fe,thecombinedanalysisof26Aland60Fe 1996; Plu¨schkeetal. 2000; Cervinõetal. 2000; Plu¨schkeetal. decaysignalsappearedasapotentiallystrongconstraintonthe 2002; Kno¨dlsederetal. 2002) and the latest studies agree that processes that govern the nucleosynthesis in massive stars and the level of the 1809keV emission from Cygnus as deduced SNe. from CGRO/COMPTEL observations is a factor of 2-3 above The launch in 2002 of the INTEGRAL gamma-ray obser- the predictions, when the latter are based on the theoretical vatory,equippedwiththeSPIhigh-resolutionspectrometer,has yields by Meynetetal. (1997), Woosley&Weaver (1995) and openedthe possibility ofextendingthe analysisof the gamma- Woosleyetal.(1995)andontheobservedstellarcontentofthe ray line emission from the Galaxy and add spectrometric in- Cygnusregion(includingtheupwardrevisionofthestellarpop- formation to the COMPTEL results. Both the 26Al and 60Fe ulationofCygOB2byKno¨dlseder2000).Plu¨schkeetal.(2000) decay signals have now been unambiguously detected by SPI show that the discrepancy could be alleviated if one includes ontheGalacticscale (Diehletal. 2006;Wangetal.2007).The the enhanced yields from massive close binary systems in the stronger26Alemissionmadeitpossibletoanalyseitsmorphol- calculation. Alternatively, Plu¨schkeetal. (2002) argue that the ogyinmoredetailandappearedparticularlyprominentinsome problem may well come from an underestimate of the stellar specific star-forming regions of the Galaxy, in agreement with content of the Cygnus OB associations due to a strong inter- the COMPTEL observations. In a previous work (Martinetal. stellarextinction,aswasshowntobethecaseforCygOB2.In 2009,hereafterPaperI),weshowedthattheCygnusregionex- Kno¨dlsederetal.(2002),theauthorsspeculatethatfuturestellar hibitsarelativelystrongsignalof26Aldecay,soweinvestigated models, especially those that include the effects of stellar rota- furtherthegamma-raylineemissionofboth26Aland60Fefrom tion,couldsolvethedilemma. Cygnus. InPaperI,weusedtheINTEGRAL/SPIobservationstore- The Cygnus region and its peculiarities were presented in assess the 26Al and 60Fe decay emission from the Cygnus re- detailin PaperI and we onlyrepeatthe main characteristicsof gion.In particular,we disentangledin a coherentway the con- the region here and the principal arguments for studying it in tribution of the nearby clustered OB population of the Cygnus nuclear gamma-ray lines. The Cygnus region, which we con- complex from the contribution of the more diffuse and spa- sider to be that part of the Galactic disk located between lon- tiallyextendednon-clusteredpopulation,whichislikelytocon- gitudes 70◦ and 90◦, harbours a very impressive concentration tribute as well to the observed emission. We now compare our of massive stars at only ∼ 1-2kpc from us. About 170 O stars INTEGRAL/SPI observations of the gamma-ray line emission have been recorded (Kno¨dlsederetal. 2002), distributed in 6 oftheCygnuscomplexwiththeoreticalpredictionsbasedonthe OB associations and a dozen open clusters, among which the latest models of stellar nucleosynthesis.In recent years, stellar spectacular Cyg OB2 cluster that alone comprises 100-120 O modelshavebeenimprovedwithmodifications,suchasthein- stars(Kno¨dlseder2000;Comerońetal.2002).Inthefollowing, clusion of stellar rotation or the revision of the mass loss rates wecollectivelyrefertotheseOBassociationsandopenclusters andnuclearcross-sections.Theseupgradeshavealreadyproven (the details of which are presented later) as the “Cygnus com- to bring simulations closer to observations. For instance, mod- plex”. With this term, we aim at gathering all these rich and els includingrotationcan accountforchangesin surfaceabun- nearby stellar groups that are relatively correlated in terms of dancesoccurringduringthemainsequencephases(Hunteretal. distance(between1and2kpc)andthuscontributetothespeci- 2008;Maederetal.2009)andfortheobservedvariationsofthe ficityoftheCygnusregion.This“Cygnuscomplex”appearsto numberofWolf-RayettoO-typestarsasafunctionofmetallic- be quite young, as shown by the lack of radio supernova rem- ityinconstantstarformationregions(Meynet&Maeder2005). nants (SNRs) or pulsars. The Green catalogue1 gives about 10 Inthefollowing,weshowthatthenewstellarmodels,combined radio SNRs between 70◦ and 90◦, but according to a compila- withtheworkdoneonobservationsinPaperI,considerablyal- tion of distances by Kaplanetal. (2004), none of these objects leviate the long standing discrepancy between theoretical and can be related to the Cygnus complex. Wendkeretal. (1991) observeddecayfluxesfromCygnus. concluded from the sensitivity of their radio survey of the region thatno other radioSNR exists up to a distance of 10kpc. 2. ObservationalconstraintsfromINTEGRAL/SPI Additionalevidencefortherelativeyouthofthecomplexcomes from the absence of radio pulsars towardsCygnus up to 4kpc, In PaperI, we presented the analysis of about 4 years of as indicated by the ATNF catalogue2 (Manchesteretal. 2005), INTEGRAL/SPIobservationsamountingtoatotaleffectiveex- althoughrecentobservationswiththeLATtelescopeaboardthe posuretimeof63.2Ms.Mostofthisexposureisconfinedtothe FermiGamma-RaySpaceTelescopehaveledtothediscoveryof Galactic plane, with 10.8Ms specifically covering the Cygnus theyoungenergeticpulsarPSRJ2032+4127(Abdoetal.2009) region.Inthissection,wereviewthemainresultsthatwereob- thatislikelyassociatedwithCygOB2(Camiloetal.2009).For tainedonthe26Aland60Fedecayemission. thosereasons,theCygnusregionappearswell-suitedtostudying massivestarsintheirhydrostaticphases. 2.1.The1809keVemissionfrom26Al The Cygnus population of massive stars is close and rich enough to ensure sufficient gamma-ray line fluxes and there- The 26Al isotope is produced predominantly by massive stars fore constitutes a good opportunity to test our understand- at various stages of their existence and it is released in the ing of the structure and evolution of massive stars, as re- ISM by both the stellar winds (of Wolf-Rayet stars espe- vealed by their nucleosynthetic activity. A number of stud- cially) and the SNe. Nucleosynthesis of 26Al follows from the 25Mg(p,γ)26Alreaction,whichmostly occursduringH central- 1 GreenD.A.,2006,”ACatalogueofGalacticSupernovaRemnants burning(with25Mgcomingfrominitialmetallicity),C/Neshell- (2006 April version)”, Astrophysics Group, Cavendish Laboratory, burning (with 25Mg coming from the CNO initial metallic- Cambridge,UK(http://www.mrao.cam.ac.uk/surveys/snrs/). ity through the 14N(α,γ)18F(e−)18O(α,γ)22Ne(α,n)25Mg chain) 2 http://www.atnf.csiro.au/research/pulsar/psrcat/ and explosive burning of the C-shell (with 25Mg coming from PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex 3 24Mg(n,γ)25Mg,24MgbeingaproductofC/Neburning). wellbethatitsstellarpopulationhassofarnotproducedasingle The mean 26Al lifetime of about 1Myr allows it to diffuse SN, which would meanno release of 60Fe into the ISM, hence inthelocalISMbeforedecayinginto26Mg.Asaconsequence, no1173/1332keVsignalfromtheCygnuscomplexatall.Asa the corresponding 1809keV emission is of diffuse nature with matteroffact,wedidnotobserveany1173/1332keVemission an overall distribution that follows that of massive stars in our fromthe Cygnusdirectionandderiveda 2σupperlimiton the Galaxy, as also traced for instance by the microwave free-free 60Fe decay flux of 1.6 × 10−5phcm−2s−1 (consistent with the emissionfromHIIregionsortheinfraredemissionfromheated valueobtainedbyWangetal.2007). dust (Kno¨dlsederetal. 1999). Most of the Galactic 1809keV flux is thus confined to the Galactic plane and inside the inner regions,exceptforanotableemissionfromtheCygnusandSco- 3. Populationsynthesis Cenregions(seePaperI). In this section we present the most important aspects of the FromtheINTEGRAL/SPIdata,wehavefoundthatthedif- populationsynthesissimulationswe performedto compareour fuse 1809keV emission from the Cygnus region is well repre- INTEGRAL/SPI observations with. In this effort, we took ad- sented by a 3◦ × 3◦ 2D Gaussian intensity distribution (which vantageoftherecentprogressbytworesearchgroupsinthefield implies a characteristic angular size of 9-10◦) centred approxi- ofmassivestarsnucleosynthesis.Inthefollowing,weintroduce matelyonthepositionoftheCygOB2cluster.Thisconfirmsex- thespecificitiesofeachgridofstellarmodelsandprovidesome pectationsthatthelatterdominatestheenergeticsandnucleosyn- detailabouttheirimplementationinourpredictivetool. thetic activity in this area. It should be emphasised, however, thatweonlyhaveweakconstraintsonthemaximalextentofthe emissionduetointrinsiclimitationsofthecodedmaskimaging 3.1.Thestellarmodels systemoftheSPIinstrumentandyet,thesizewefoundisconsis- Once the massive stars had been recognised as the dom- tentwiththeresultsobtainedfromtheCGRO/COMPTELobser- inant source of the Galactic 26Al (Prantzos&Diehl 1996; vations(Plu¨schkeetal.2001).Thetotalfluxfromtheextended sourceintheCygnusdirectionis(6.0±1.0)×10−5phcm−2s−1, Kno¨dlsederetal.1999),thequestionofitsoriginshiftedtoiden- where(3.9±1.1)×10−5phcm−2s−1isattributedtotheCygnus tifyingthesubsetofmassivestarsand/orevolutionaryphasesre- sponsible for its ejection. For historical reasons, mostly linked complexandthe remainingflux stemsfromGalactic diskfore- totheevolutionofthestellarmodels,two“competing”sources groundandbackgroundemissions. were discussed: stars less massive than about 35M through ThankstothehighspectralresolutionoftheSPIinstrument, ⊙ their SNII explosions, and stars more massive than 35M we were able to study the profile of the 1809keV line. We ob- ⊙ through their WR winds (see the review by Limongi&Chieffi served a line position corresponding to a radial velocity of 33 ±66kms−1,whichagreeswiththeoverallradialmotionsinthe 2006, hereafter LC06). The debate then settled on the relative contributionofSNIIandWRtotheGalactic26Albudgetand,in direction and at the distance of the Cygnus complex, as indi- this situation, the 60Fe appeared as a promising discriminating catedforinstancebyspectrometricobservationsoftheCO line agentsinceitisreleasedonlybySNexplosions. (Dameetal. 2001). We also measured a slight line broadening that,withinthestatisticaluncertainty,isconsistentwith26Albe- TheworkofLC06hasbroadenedtheissuethankstoanew gridofstellarmodels.Theauthorscomputedthenucleosynthe- ingstillorinamediumwithturbulentorexpansionvelocitiesof upto100-200kms−1. sisyieldsforawiderangeofnon-rotatingsolarmetallicitymod- els, with initial stellar masses from 11 to 120M . The models ⊙ were followed over their entire life, including both the hydro- 2.2.The1173and1332keVemissionfrom60Fe static phases and the explosive burning episodes of the super- novaexplosion.ThemainoutcomesoftheLC06studyare: Themassivestaroriginofthe60Feisotopeisfarlessestablished than for 26Al. The nucleosynthesis of 60Fe by massive stars 1. Overthemassintervalconsidered,mostoftheejected26Al wasaddressedbyTimmesetal.(1995)andextensivelyreviewed has explosive origins. For the particular case of the most by(Limongi&Chieffi 2006). Fromthese works, we knowthat massivestars, typicallytheWR progenitors,the production massivestarsareasourceof60Fethroughthe59Fe(n,γ)60Fere- of26AlduetoexplosiveburningoftheC/Neshellsexceeds actionthatoccursduringHeshell-burning,Cshell-burning,and the quantity synthesised by central H-burning and carried explosive burning of the C-shell. In these processes, the seed awaybythestellarwinds. amounts of Fe come from the initial metallicity and so do the 2. The productionof 60Fe (by C/Ne convectiveshells burning neutrons,throughthe 22Ne(α,n)25Mg reaction (with 22Ne com- mostly)of themostmassivestarsthatexplodeasSNIb/cis ing from the initial CNO, see 2.1). The 60Fe thus producedby up to a factor 10 higher than the yields associated with the massive stars is then released in the ISM only through SN ex- lowermassstarsthatexplodeasSNII. plosions and no wind contribution is expected. Other objects such as AGB stars (Karakas&Lattanzio 2007) or SNe of type IthasalwaysbeenclearthatadichotomybetweenWRandSNII Ia (Iwamotoetal. 1999) are also potential producers of 60Fe. asthesourcesof26Alisinadequateandthatthewholecontribu- Although the relative contribution of each class of objects re- tionofeachmassivestarshouldbeaccountedfortounderstand mains entirely unconstrained on the observational level, the theoriginofboththe26Aland60Fe.Moreover,theinitialmetal- available models for the various potential sources indicate that licity mayhavea strongimpactontherelativecontributionsof core-collapseSNeshouldprovidemostoftheGalactic60Fe. WRwindsandSNexplosionstotheproductionof26Al,aswill The observed intensity of the Galactic decay emission at beillustratedlater. 1173 and 1332keV is a factor of 7-8 lower than the 1809keV In parallel to the work of LC06, grids of stellar models emission from 26Al (Harrisetal. 2005; Wangetal. 2007, and including stellar rotation have revealed that the latter strongly PaperI), so the detection of any 60Fe signal from Cygnusmay affects the evolution and nucleosynthesis of massive stars. atfirstsightbedifficultgiventhesensitivityofINTEGRAL/SPI. Meynet&Maeder(2003,2005)havecomputedtheevolutionof Inaddition,theCygnuscomplexseemsquiteyoung,anditmay rotating modelsof stars with initial masses from20 to 120M ⊙ 4 PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex 3.2.Thegridofyields The LC06 grid is complete and homogeneousbut does not include stellar rotationand is onlyavailable for solar metallicity, while the MMP05 grid includes the effects of stellar rotation and metallicity, but only covers the longest hydrostatic phases of those stars initially more massive than 20M . To make the ⊙ most realistic predictionsof the nucleosynthesisactivity of the Cygnuscomplex,wethereforebuiltourowngridofyieldsfrom theworksofLC06andMMP05. Before entering the description of our grid, we emphasise theterminologythatwillbeusedthereafter.Wedividedthenu- cleosynthesiscontributionsinto“windyields”and“SNyields”, whichis notequivalentto a separationinto hydrostaticandexplosive yields. Wind yields pertain to 26Al alone and are obvi- ouslyofhydrostaticorigin,whereasSNyieldsconcernbothiso- topes and have both hydrostatic and explosive origins. For in- Fig.1.Supernovayieldsof26Aland60FeasafunctionoftheCO stance,the60FereleasedinaSNexplosioncomesfromthehy- drostaticburningoftheHeandCconvectiveshellsbutalsofrom coremass(fromthemodelsofLC06).Thesecurvesareusedto theexplosiveburningoftheCshell.Wealsonotethatthewind associate the purelyhydrostaticmodelsof MMP05with super- yieldisaprogressive,time-dependentoutflowof26Al,whereas novayields. theSNyieldisconsideredaninstantaneousreleaseof26Aland 60Fe. Thegridweused inourpopulationsynthesiscodeisbased onthefollowingassumptions: uptotheendofcentralHe-burning.Theserotatingmodelswere computedforvelocitiescompatiblewiththemeanobservedve- 1. Thelowermassstars,from11upto20M ,arerepresented ⊙ locitiesforOBstars,andtheimpactofstellarrotationonthepro- by the models of LC06 for both their wind and SN contri- ductionof26AlhasbeenexaminedbyPalaciosetal.(2005).The butions3. It has also been assumed that these LC06 models mainoutcomesoftheseworks(hereaftercollectivelyreferredto arevalidatmetallicitiesotherthansolar.Wewilldiscussthis asMMP05)are: assumptionlateron. 2. The higher mass stars, from 20 to 120M (typically those ⊙ 1. Rotation increases the convection and the circulation of thatbecomeWRstars),arerepresentedfortheirwindcontri- chemicalspeciesinsidethestar.Thisimpliesalargerinitial butionbytherotatingmodelsofMMP05forthefourmetal- reservoir of 25Mg available for proton capture (in the case licities considered by these authors. Each of these stellar ofnon-zeroinitialmetallicity)andanearlierejectionofthe tracks is then associated to one of the SN yields of LC06 26AlproducedbycentralH-burning. throughitsCOcoremass(asexplainedbelow). 2. Rotationincreasesthemasslossbyalteringthesurfacegrav- ity,theeffectivetemperature,andtheopacityoftheenvelope. Inthecompositionofourgrid,weassumedthattheSNyieldis This,combinedwiththemoreefficientdiffusionoftheburn- strongly linked to the final CO core mass of the model (a hy- ingproductstothestellarsurface,resultsinanearliertransi- pothesisalsousedbyCervinõetal.2000).Thiscanfirstbejus- tiontotheWRphaseandadecreaseintheminimuminitial tified by the fact that, at the moment of core-collapse, most of mass to becomeWR. As a result, the 26Al containedin the the26Aland60FesynthesisedsofarislockedintheCOcoreas Hecore(followingcentralHburning)isreleasedsoonerand aresultoftheHe/C/Neshell-burningepisodes.Second,itseems bymorestars. reasonable to consider that the amounts of 26Al and 60Fe pro- ducedexplosively(throughexplosiveburningoftheC-shell)de- In summary,stellar rotation alters the 26Al yields in two ways: pendonthesizeoftheCOcore(evenifotherparameters,such thequantitiesejectedintheWRwindsareincreasedandthetime asthechemicalprofileandthemass-radiusrelation,maybeim- profile of the injection into the ISM is differentas well. While portant). From the LC06 grid of models, it is possible to draw thedetailsofthe 26Alreleasewith timeare notimportantfora relations between the SN yields and the CO core masses (see large,stationarysystem such asthe Galaxy,forwhichonlythe Fig.1).Theserelationsarethenusedtoassociatethepurelyhy- totalyieldsperstararerelevant,theCygnuscomplexofmassive drostatic modelsof MMP05 with SN yields throughtheir final starsisayoung,evolvingregionandthecomparisonofobserva- COcoremassthat,forsameinitialstellarmass,differsfromthat tionswithpredictionsthereforeincludesatemporaldimension. obtainedby LC06. Indeed,forstars belowabout60M , stellar ⊙ Therefore,areliablepopulationsynthesismodelshouldnotonly rotationincreasesthefinalCOcoremassbecauseofanenhanced be able to reproduce the observed 1809keV flux from Cygnus convection;above60M , rotationcauses a decrease in the CO ⊙ butalso reach an agreementwithin a time rangethat is consis- coremassbecauseofanenhancedmassloss(Meynet&Maeder tentwiththeageestimatesofthespecificstellarclusters. 2003,2005). Besidestheaboveimprovements,bothgridsofstellarmod- Thisapproachtakesintoaccounthowstellarrotationcanaf- elsarealsobasedonrevisedmasslossratestakingtheeffectof fecttheSNyieldsthroughitsimpactonthestellarevolution.It clumping in the winds into account and on revised nuclear re- should still be emphasised that the SN yields used in our grid action rates. MMP05 also explored the influence of the initial metallicityZ bycomputingfourgridsatZ = 0.004,0.008,0.02, 3 In practice, however, the 26Al wind contribution of these stars is and 0.04. The grid of LC06 was only computed for a Z=0.02 negligible because they experience little mass loss compared to the solarmetallicity. highermassstars. PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex 5 Fig.2.26AlwindyieldsfromLC06,MMP05,andWH07,com- Fig.3.26AlsupernovayieldsfromLC06andWH07,compared paredtotheolderresultsfromMM97. totheolderresultsfromWW95.Alsoplottedaretheyieldsfor MM97 and MMP05 extrapolated through CO core mass from WLW95andLC06respectively. come only from non-rotatingmodels. The impactof stellar rotation on the nucleosynthesis of the latest evolutionary stages remainsunexploreduptonow.LC06pointedouthowimportant lar winds. The WH07 wind yields are all above the yields of the shell convection efficiency is to the synthesis of 26Al and LC06(uptoafactorof5insomecases)butalwaysremainbe- 60Fe, but how stellar rotationaffects the nucleosynthesisof the low the MMP05 yields. This fact very likely comes from the He/C/Ne shell-burning episodes is still an open question. Our absenceofrotationintheWH07models,butthedifferencebe- limitedknowledgeonthatpointthereforeconstitutesapotential tweenWH07andLC06couldseemsurprising.Indeed,alower sourceoferrorforourSNyields. initialmetallicity like Z=0.016meansthatall otherparameters In Figs. 2 and 3, we show the 26Al yields for LC06 and being unchanged, a reduced initial reservoir of 25Mg available MMP05 (where the SN yields for MMP05 were extrapolated forprotoncapture,hencelower26Alwindyields,butathorough through CO core mass, as explained above), compared to the comparisonoftheWH07andLC06modelsisbeyondthescope previous generation of yields from Meynetetal. (1997, here- of this paper. The same rates were used for the main reactions after MM97),Woosley&Weaver (1995, hereafterWW95) and governing26Alproduction,butthehighermassmodelsofWH07 Woosleyetal. (1995, hereafter WLW95). This previous gener- werefoundtoexperiencemoremassloss(asdiscussedinmore ation of yields was used in a population synthesis analysis by detailbelow).Thismayexplainthe situationabove40M . For ⊙ Cervinõetal. (2000) and Kno¨dlsederetal. (2002), who con- the lower-mass models, however, the higher 26Al wind yields cluded that they could not account for the 1809keV flux from obtainedwithalowerinitialmetallicitymaystemfromthedif- CygnusobservedbyCGRO/COMPTEL(seealsoPlu¨schkeetal. ferenttreatmentoftheconvectivephysicsand/ortothedifferent 2001).InFig.4,weshowthe60FeyieldsforLC06andMMP05 mass loss rate prescriptionduring main sequence or red super- (where again the SN yields for MMP05 were extrapolated giantstages. throughCOcoremass). In Fig. 3, the 26Al SN yields of LC06 are plotted together In all three figures, we also compare the set of yields ob- with the SN yields extrapolated through the CO core mass for tained by Woosley&Heger (2007, hereafter WH07). These theMMP05rotatingWRmodels.Oneclearlyseestheeffectof yields come from a large grid of stellar models (from 13 to rotationonthesizeoftheCOcore,henceontheextrapolatedSN 120M ) that basically includes revised mass loss rates and re- yields, asdiscussed above.An oldergrid ofyieldsis also plot- ⊙ vised nuclear reaction rates, but no rotation. The authors used ted,whichconsistedoftheWW95SNIIyieldsupto25M and ⊙ an initial metallicity of Z=0.016, corresponding to the revised the SNIb/c yields associated with the MM97 non-rotating WR solarabundancesofLodders(2003),andcomputedbothhydro- models.FortheseSNIb/cyields,theyieldversusCOcoremass static and explosive phases. In its general features, the WH07 relationwas derivedfromthe pureHe star modelsofWLW95. grid is very similar to the LC06 grid, except for the different Regardingthe26AlSNyieldsofWH07,theyareverysimilarto initialmetallicity.Inthedetails,however,manyaspectsofstel- thoseofLC06upto40M (variationsremainwithin±50%)ex- ⊙ larphysicswereimplementeddifferently(suchasthemassloss ceptforthestarsaround14-15M ,forwhichLC06foundadra- ⊙ prescriptions,seebelow).Asseeninthenextparagraphs,thedif- maticincreaseofthe26AlSNyieldnotseeninWH07.Beyond ferencesbetween the WH07 yields and the LC06 and MMP05 40M ,theyieldsofWH07arebelowthoseofLC06byafactor ⊙ yields illustrate the extent to which the implementation of the of up to 10 for the 120M model. This is very likely an effect ⊙ stellarphysicscanaffectthenucleosynthesisresults. ofthestrongermasslossexperiencedbytheWH07highermass In Fig. 2, we see that the MMP05 yields obtainedfrom ro- models. tatingmodelsaremoreenhancedthantheyieldsofMM97,and Again, no detailed comparison was performed but we thisdespitethatthemasslossratesusedintheMMP05models notedthattheWellstein&Langer(1999)masslossprescription arelowerthanintheMM97models.Thisillustratestheimpor- (scaled by 1/3 to account for clumping) used by WH07 gives tanceoftheeffectsofrotationthat,asexplainedin3.1,favours masslossratesinthelateWRstages(typicallyWNEandWCO the production of 26Al and its subsequent ejection by the stel- stars)thatareafactorof2-3abovetheNugis&Lamers(2000) 6 PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex Cygnus region have fostered a great number of studies that togetherprovideuswithaprettygoodknowledgeofthearea.In particular,recentworkshaveleadtoasubstantialrevisionofthe stellar contentof the Cyg OB2 cluster and thusallowfor more accuratepredictions. From the stellar census of Kno¨dlseder (2000), Kno¨dlsederetal. (2002), and LeDuigou&Kno¨dlseder(2002), weselectedthemostrelevantOBassociationsandopenclusters of the Cygnus region. We restricted ourselves to a longitude range of 70-90◦ and among the stellar groups falling in that interval, we excluded the ones with ages above 20Myr (since their decay fluxes are negligible after that time, see Fig. 6). The selected objects are listed in Table 1, together with their maincharacteristics.Kno¨dlsederetal.(2002)indicatethatmost open clusters are potentially physically related to some of the OB associations. In total, our selection amountsto a total of ∼ 170 O stars, most of which are in the Cyg OB2 cluster. This Fig.4.60FesupernovayieldsfromLC06andWH07.Alsoplot- ensemble is what we have called so far the Cygnus complex. ted are the yields for MMP05 extrapolated through CO core Schneideretal.(2006)demonstratethatCygOB1,2,and9very massfromLC06. likely originated from the same giant molecular cloud, which givesaphysicalrealitytothisgrouping.TheconnectionofCyg OB3,7,and8tothiscomplexis,however,stilluncertain. Whether all these clusters share a common origin does not prescriptionusedbyLC06.Thisisobviousfromthefinalpresu- matteratallforourpurpose.Ouraimhereissimplytogatherthe pernovamassesofthe40-120M⊙ models:whilethoseobtained nearbyclusteredstellarpopulationthatisverylikelyresponsible by LC06 are in the 12-20M⊙ range, those obtained by WH07 formostofthestrong1809keVemissionfromCygnus.Insome are between 6-10M⊙. It is interesting to note that LC06 tested previous studies (delRioetal. 1996; Plu¨schkeetal. 2002), the the impact of increased mass loss during the WNE and WCO total 1809keV flux from the Cygnus region was compared to phasesbyusingtheprescriptionofLanger(1989),whichgives theoreticalpredictionsbased on the clustered stellar contentof ratesafactorof2-3abovetheratesofNugis&Lamers(2000). the Cygnus region supplemented by a few isolated WR stars Inthiscase,their40-120M⊙modelsendedupwithfinalmasses and SNRs. In our study, we focus both observationally and and 26Al SN yields that are very similar to those of WH07. A theoretically on the clustered stellar contentalone, because the stronger mass loss actually reduces the size of the He and CO clustered population is likely to be better inventoried than the coresandleadstolower26AlSNyields.Thedifferencebetween isolated or non-clustered population. In PaperI, we used the the WH07 and LC06 26Al SN yields is all the more important INTEGRAL/SPI data to disentangle the 1809keV flux of the becausein the WH07 yieldsthere is a contributionfromthe ν- Cygnus complex from the more spatially extended emission processthatisabsentintheLC06yields(see4.1.3).Thealterna- likely due to unclustered objects like non-member WR and tivegridofWH07thusillustrateshowimportantmasslosscan OB stars (runaway stars or stars from previous star formation befornucleosynthesis. episodes, see Comeroń&Pasquali (2007) and Comerońetal. InFig.4,the60FeyieldsofLC06andWH07areplottedwith (2008)) or isolated SNRs. In the present paper, we compare the SN yields extrapolated through the CO core mass for the this flux to population synthesis predictions based only on the MMP05rotatingWR models.Asfor26AlSN yields,the effect Cygnuscomplexpopulation. ofrotationonthesizeoftheCOcore,henceontheextrapolated yields, is manifest. The WH07 yields are apparently quite dif- 3.4.Thecode ferentfromthoseofLC06. Between10-20M theyareuptoa ⊙ factorof20abovethoseofLC06,thenbetween20-50M⊙ they Ourpopulationsynthesiscodeisverysimilartomanyothersuch arestillabovethoseofLC06bymoremoderatefactorsofupto tools (like Leithereretal. 1992; Cervinõetal. 2000; Vossetal. 5,andbeyond50M⊙ theydropbelowLC06byquitelargefac- 2009,forexample).Forthecaseweareinterestedin,theparam- tors.Forthehighermassmodels,theloweryieldsofWH07very etersofasimulationarethecharacteristicsofthestellarpopula- likelycomefromthestrongermasslossofWRstars,asalready tionto be modelled:slope andboundsof the IMF(initialmass discussedabove.Hereagain,thesameeffectwasnotedbyLC06 function, assumed here to be a homogeneouspower-law func- whentheyusedtheincreasedmasslossratesofLanger(1989). tion),stellarcontent(givenasthenumberofstarsobservedina For the lower masses, however,the differencesare not so easy certainmassinterval),andmeandistancetothecluster(tocom- toexplain.Thelowerinitialmetallicityshouldhaveresulted,all putethedecayfluxes).Weassumethatthestarformationoccurs otherparametersbeingunchanged,inloweryieldssince60Feisa asaninstantaneousburst,whichmeansthatallstarsformatthe puresecondaryproduct.Themajorreactionratesfortheproduc- sametime(discussedlater). tionanddestructionof60FewerefoundtobeidenticalinWH07 Oncethepopulationtobesynthesisedhasbeendefined,we and LC06, so the difference may come from another aspect of create a cluster realisation by randomly sampling the IMF un- stellarphysicssuchasconvectionormixing. til the observationalcriterion (the numberof observedstars) is met.Then,allstarsareevolvedintimestepstypicallyof104yrs. For each run, we compute the time-dependent release of 26Al 3.3.ThestellarcontentoftheCygnuscomplex and60Fe,thedecayfluxesat1809and1173/1332keVandafew The last ingredient of our simulation is the definition of the otheroutputssuchasthemechanicalluminosity.Toevaluatethe stellar population to be modelled. The peculiarities of the effectsoffinitesampling,agivenstellarpopulationismodelled PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex 7 Table1.CharacteristicsoftheCygnusOBassociationsandstel- the13or16M models.Asaconsequence,itshouldberemem- ⊙ larclusterscollectivelyreferredtoastheCygnuscomplex(from beredthatthe finite grid ofyieldsand the associated interpola- Kno¨dlsederetal.2002). tionschemearepotentialsourcesoferrors. Name Observedpopulation Distance(pc) Age(Myr) 4. Predictionsversusobservations CygOB1 23∈[15,40M ] 1905 4 ⊙ BasedonthecodeandinputspresentedinSect.3,wehavesim- CygOB2 120∈[20,120M⊙] 1584 2.5 ulatedthehistoryoftheejectionof26Aland60FeintheCygnus complexandcomputedthecorrespondingdecayfluxes.Wefirst CygOB3 14∈[25,60M ] 2187 3.5 ⊙ focused on the purely photometric aspect, trying to reproduce CygOB7 10∈[7,25M⊙] 832 3.5 theobservedfluxesbysimplyaddingthecontributionsfromall the stellar groupslisted in Table 1. In the second part, we also CygOB8 6∈[20,40M ] 2399 7.5 ⊙ attemptedtoreproducethespatialextentofthisemission. CygOB9 8∈[20,40M ] 1259 3.5 ⊙ Ber86 11∈[7,25M ] 1660 4 4.1.Thedecayemissionfluxes ⊙ Ber87 24∈[7,25M ] 1905 4.5 All the stellar groups listed in Table 1 have been modelled in- ⊙ dependently,assumingacoevalstarformationforeachofthem. NGC6871 13∈[7,25M ] 2399 5.5 ⊙ Then, the lightcurves at 1809 and 1173/1332keV were added, NGC6913 13∈[7,40M ] 1820 3.5 with the appropriate time shifts to take the differences in age ⊙ of the different stellar populations into account. The resulting NGC6910 7∈[7,25M⊙] 1820 4.5 lightcurves for the whole Cygnus complex are given with t=0 correspondingtothepresentdayandthusallowedexploringthe NGC6883 2∈[12,15M ] 1820 15 ⊙ pastandfutureevolutionofthedecayfluxes. IC4996 6∈[6,25M⊙] 1660 5.5 Asafirststep,wefocusedontheresultsobtainedforanini- tialsolarmetallicity,firstlybecausemostofthepastpopulation synthesisstudieswerebasedonnucleosynthesisyieldsfromso- larmetallicitystellarmodelssoacomparisoncouldbemade,but severaltimes(typically100-200)untilthemeanstellarmassand alsobecausetheCygnuscomplexliesonthesolarcircleandis numberofstarsobtainedoveralltrialsarewithin1%ofthethe- thusexpectedtohaveanearlysolarmetallicity. oreticalvaluesforthespecifiedIMF. Foreachquantity,thisal- lows computing the mean value and the statistical variance at eachtimestepinthelifeofthecluster. 4.1.1. ThecaseforsolarmetallicityZ=0.02 Althoughasyntheticstar cantakeanyinitialmassbetween The results obtained for an initial solar metallicity are pre- 11 and 120M (the boundsof our grid of yields), we have the ⊙ sentedinFig.5(top),togetherwiththeobservationalconstraints nucleosynthesis yields for only a finite number of initial stel- we derived from the INTEGRAL/SPI observations. The pre- lar masses, typically 11, 12, 13, 14, 15, 16, 17, 20, 40, 60, 80, dicted decay fluxes at 1809 and 1173/1332keV are there to- 120M (the MMP05 grids of the higher mass models differ ⊙ gether with their typical variances due to the finite IMF sam- for the various metallicities). We therefore need an interpola- pling. From the open clusters and OB associations ages deter- tionschemetocomputetheyieldsofthesyntheticstarsfromthe mined by Kno¨dlsederetal. (2002), our current position along onesatourdisposal. the time axis is indicated by the zero point. Yet, these clus- The characteristicsof a synthetic star with a randominitial ter ages were determined by fitting isochrones to Hertzprung- mass are determined from the closest lower and higher mass Russeldiagrams,andtheisochronesusedbytheseauthorswere models (for instance, the characteristics of a 53.6M star are ⊙ builtfromthenon-rotatingstellarmodelsavailableatthattime. determinedfromthe40and60M models).Afirstcubic-spline ⊙ Rotationisnowknowntoincreasethelifetimeofthemodelsby interpolationisperformedtodeterminethedurationsofthemain 15-25%, mainly through augmenting the main-sequence dura- evolutionary sequences of the star: central H-burning, central tion(Meynet&Maeder2003).Therefore,theclusterageshave He-burningandfinalstages(whenavailable).Then,foreachof probablybeenunderestimatedbythesameamount,andourcur- thesethreesequences,the26Aland60Fequantitiesejectedovera rent position on the time axis of Fig. 5 is likely shifted to the giventimefractionofthesequenceareobtainedbylinearinter- rightofthezero-point,asindicatedbythegreyshadedarea. polationofthequantitiesejectedbytheclosestlowerandhigher Insidethistimerange,thereisclearagreementbetweenthe mass models over the same time fraction. When the synthetic predicted1809keVfluxfrom26Alandtheobservedvalue,while starexplodes,its26Aland60Feyieldsaredeterminedbyalinear the theoretical level of the 60Fe decay emission is consistent interpolationbasedontheCOcoremassandreleasedinthecur- withourupperlimit.Ourpopulationsynthesisindicatesthatwe renttimestep. are currentlyin a periodofsteep increase of the1809keV flux The work of LC06 clearly shows that the yields are not (overMyrtimescales)thatfollowsfromasubstantialreleaseof smooth functions of the initial stellar mass4. For instance, the 26Al through the stellar winds of the most massive stars, sup- lackofaCconvectiveshellintheir14and15M modelstrans- ⊙ plementedbythefirstsupernovaexplosionsthatoccurredinthe latesintoadramaticincreaseinthe26Alproductioncomparedto oldest stellar groups, typically those above 3.5Myr (the short- 4 Onthecontrary,thedurationsofthemainburningstagesareaquite est lifetime in our grid of models, corresponding to an initial smoothfunctionoftheinitialmass,closetoapower-law,andthatiswhy 120M⊙ star).Atabout1Myronthetime axisofFig. 5(which inourworktheyaredeterminedbyacubic-splineinterpolationinstead may be our current position), most of the wind 26Al has been ofasimplelinearinterpolation. injectedintheISM;thesupernovaethentakeoverandnowre- 8 PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex lease both 26Al and 60Fe. From there on, the 1809keV flux is showthatsomefractionoftheCOmassobservedtowardCygnus expectedto rise again,less rapidlythough,andto reachwithin isactuallylocatedbehindtheOBassociations,whichmeansthat ∼ 1Myr a maximum mean value of ∼ 4.2 × 10−5phcm−2s−1, theappliedcorrectionwasincorrectinsomecases. which remains in our measured flux interval. Regarding the Our work indicates that the past dilemma resulted from an 1173/1332keV emission, the most massive stars of the oldest overestimateofthe1809keVfluxoftheCygnuscomplex,com- stellar groupsexplodedand consequentlystartedinjecting60Fe binedwithanunderestimateofthenucleosynthesisyields.First intheISM.Thenumberofeventsuptonow,however,istoolow of all, in PaperI, a careful analysis of the INTEGRAL/SPI tohavebuiltupadetectableamountof60Fe.Moreover,ourpre- 1809keV observations showed that the decay flux attributable dictionshowsthateventhemaximum1173/1332keVflux,tobe to the Cygnuscomplexis only ∼ 65%of the total flux coming reachedin∼2-3Myrfromnow,remainsbelowourpresentup- fromtheCygnusregion.Thisresultwasobtainedbyseparating perlimit. the emission due to the Cygnus complex from the foreground Our simulation predicts that a certain number of SNe, typ- andbackgroundmeanGalactic contribution.Then,stellar rota- ically 10-20 over the last Myr, exploded in the Cygnus com- tion, as discussed in Sect. 3.1, turned out to be a key factor in plex,whichcontradictstheapparentabsenceofradioSNRsand nucleosynthesissinceitenhancestheproductionof26Albutalso the detection of only a single pulsar in its various OB associ- affectstherateofitsreleaseintheISM.Inaddition,thecompu- ations and open clusters. The absence of SNRs could be ex- tation in a coherentway of the SNIb/c contribution (by LC06) plainedbythefactthatSNRsexpandinginsidethehottenuous showed that the SN yields of the most massive stars cannotbe interiorsofsuperbubblesexhibitveryweak characteristicradio neglectedorsimplyextrapolatedfrompureHestarscalculations. and optical signatures (Chu 1997). With their high stellar content, the OB associations of the Cygnus complex are expected 4.1.2. ThecaseformetallicityZ=0.01 to blow large superbubblesinside of which the SNR are likely to be invisible. Moreover, the typical radio lifetime of a SNR Mostofthepaststudieswerebasedonthenucleosynthesisyields is short compared to the lifetime of 26Al (104 versus 106yrs, ofsolarmetallicitystellarmodels,sowepresentedtheabovere- Frailetal. 1994) andall SNRs may wellbe completelydiluted sultforcomparison;yet,thereareindicationsthatthestellarpop- in the ISM by now. In that case, we would expectto see some ulationoftheCygnusregionmaybesubsolarwithametallicity manifestationsoftheircompactremnants.Uptonow,onlyasin- ofaboutZ=0.01(Daflonetal.2001).Althoughwedonothavea gle pulsar has beenproposedassociated with the Cygnuscom- fullyhomogeneousgridofstellarmodelsofvariousmetallicities plex, more precisely with Cyg OB2 (Camiloetal. 2009), de- coveringhydrostaticandexplosivephasesatourdisposal,weex- spite the dedicated deep surveys that have been undertaken at ploitedasmuchaspossiblethe worksofLC06andMMP05to radiofrequencies(Janssenetal.2009).Thedifficultyoffinding estimate the 26Al and 60Fe yields of a non-solar population of pulsars in the Cygnus complex could come from an excessive massive stars. We recall here that our grid of yields was built scintillation or dispersion of the radio signals in the strongly frompurelyhydrostaticstellartracksofvariousinitialmetallic- ionised environment of such a concentration of massive stars. itiescombinedwithfullcalculations(hydrostaticandexplosive) Or it may be that the narrow cones of their radio emission do onlyatsolarmetallicity,theconnectionbeingmade(whennec- not intersect our line of sight. In the latter case, these pulsars, essary)throughtheCOcoremass(see3.2). if they exist, may become “visible” to the Fermi/LAT instru- ThedecaylightcurvesforZ=0.01(obtainedbylinearinter- ment through their gamma-ray emission, which is unabsorbed polation between Z=0.008 and Z=0.02) are presented in Fig. and less-focused compared to the radio emission. The discov- 5 (bottom). Compared to the Z=0.02 case, the peak 1809keV ery by Fermi/LAT of the gamma-ray pulsar PSR J2032+4127 flux is somewhat reduced and occurs at a later time, while the that is likely located in Cyg OB2 is promising in this respect 1173/1332keV flux is increased by almost a factor of 3. Both (Abdoetal.2009). differences come from the effect of lower metallicity on stel- Theaboveresultconstitutesastepforwardcomparedtothe larevolution.Indeed,alowerinitialmetallicityimpliesreduced previousstudiesofthenucleosynthesisactivityoftheCygnusre- masslossforstarsinitiallymoremassivethan20M ,sincethe ⊙ gion.Indeed,thelatestevaluationsofthesituationconcludethat, winds of the OB and WR phases are mostly driven by radia- from the established stellar content(includingthe revised 100- tivepressureonmetalsthroughtheirnumerousabsorptionlines. 120OstarsofCygOB2)andthestellar yieldsofMeynetetal. ReducedmasslossleadstolargerCOcoremasses,hencehigher (1997) and Woosley&Weaver (1995), the observed 1809keV SNyields,asweassumedthatthelattermostlydependsonthe flux from Cygnus is a factor of 2-3 above the predictions (see COcoremass.Thisaccountsfortheconsiderableincreaseinthe forinstanceKno¨dlsederetal.2002).Plu¨schkeetal.(2000)point 60Fe productionin the Z=0.01 case. For 26Al, however,the in- out that the possible enhanced yields of massive close binary creaseintheSNyieldsiscompensatedbyadecreaseinthewind systems could alleviate the discrepancy. While it is clear that yields;indeed,alowerinitialmetallicitymeansareducedmass such systems may occur in the Cygnus complex and could in- loss but also a reduced initial 25Mg reservoir for proton cap- deed have a strong impact on the nucleosynthesis yields, our ture(hencesynthesisof26Al)duringcentralH-burning.Onthe theoretical knowledge of this scenario is still too uncertain to whole,the1809keVlightcurveisnotdramaticallyalteredbythe allow us to take it quantitatively into consideration. Following reduced metallicity, but the stronger SN contribution explains the significant upward revision of the Cyg OB2 stellar con- the delay in the 1809keV peak flux compared to the Z=0.02 tent(Kno¨dlseder2000),Plu¨schkeetal.(2002)suggestedthatthe case. stellar content of the other OB associations of the Cygnus re- Within the likely present-day time window, the mean pre- gion maybe underestimateddue to a strong interstellar extinc- dicted 1809keV flux is at best a factor of two below the ob- tion;with correctionfactorsbased on the CO columndensities served value.Consistency is foundlater on, at ∼ 2-3Myr from in the direction of Cygnus, the authors show that the resulting now,butthenthe1173/1332keVfluxhasrisenclosetooreven increaseinthestellarcontentcouldsolvetheproblem,andyet, aboveourupperlimit.Itthereforeseemsthattheoverallagree- thisrevisionofthestellar populationhasnotbeenobservation- ment obtained at Z=0.02 does no longer holds if the subsolar ally substantiatedsince then.Moreover,Schneideretal. (2006) metallicity of Cygnus is accounted for. We may question this PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex 9 subsolar metallicity. It is indeed somewhatsurprising that stellar associationslying at aboutthe same galactocentricdistance astheSunandbeingmuchyoungercouldbesubsolar,although Daflonetal. (2001) mention that their abundances are consistent with the expected values, as predicted by various Galactic abundance gradients. While the C, N O, Si, Al, and Fe were found to be underabundant relative to solar values, Mg and S werefoundtobesolarandMghasstraightforwardrelevancefor 26Alnucleosynthesis.Theauthorsstate thatBstarsinthesolar neighbourhoodgenerallyhavesubsolarabundances,soonemay worry abouta generalsystematic bias in determiningchemical abundancesinBstars.Wealsonotethatasolarabundancewas derivedbyNajarro(2001) forthe P-Cygnistar,whichliesat ∼ 2kpc,closetotheCygnuscomplex. Fromthesearguments,wethereforeconsiderthatthemetal- licity of the Cygnuscomplexis still uncertain,especially since the associations investigated by Daflonetal. (2001) are Cyg OB3 and 7, which are not those contributing the most to the Cygnus complex. In addition to this, we discuss below several other potential sources of error or uncertainty that can mod- ifyand/oralleviatethecurrentsituationforthegamma-rayline emissionoftheCygnuscomplex. 4.1.3. Sourcesoferror/uncertainty Uptonow,wehaveonlyconsideredtheuncertaintyonthepre- dicted fluxes due to finite IMF sampling, but the uncertainty on the exact distances and ages of the stellar associations of the Cygnus complex should also be taken into account. From Kno¨dlsederetal. (2002), we estimated that around the present date, the uncertainties on the ages and distances of the stellar clusters translate into typicaluncertaintiesof ∼ 1.0 and 0.25 × 10−5phcm−2s−1 on the decay line fluxes of 26Al and 60Fe, respectively. In connection with the age of the stellar clusters, it Fig.5.Predictedversusobserveddecayfluxes,forZ=0.02(so- shouldbe notedthat ourassumption of a coevalstar formation lar) initial metallicity (top) and Z=0.01 initial metallicity (bot- favoursmaximumpeakfluxes.Themostmassivestars(between tom). The red curve corresponds to the 1809keV flux and the 30and120M )haveverysimilarlifetimes,soinacoevalpopu- ⊙ bluecurvetothe1173/1332keVflux.Thedottedlinesgivethe lationthenucleosynthesiscontributionsoftheirlaststages(WR typicalvarianceduetothefiniteIMFsampling.Theorangeband andSNe)arereleasedatalmostthesametimeandthereforelead indicatestheobserved1809keVfluxandthegreenlinetheupper to a steep rise in the decayfluxes.In contrast,if star formation limitonthe1173/1332keVemissionfromtheCygnuscomplex. occurs over a few Myrs, the nucleosynthesis contributions are Theverticalshadedareaindicatesourprobablecurrentposition spreadovera longertimeintervalandtheresultingpeakfluxes alongthetimeaxis(seetext). arelower. We emphasise once again that our grid of nucleosynthesis yields is not homogeneousand self-consistent and that the SN yields of the higher-mass stars at all metallicities were extrap- correspondingnuclearprocessesseemtodependonlymildlyon olatedfromnon-rotatingsolar models(whereasthe hydrostatic theinitialmetallicity(althoughthemanyinvolvedspeciesdonot phases of these higher mass stars were computed by including allowdrawingdefinitiveconclusions).Inthiscase, ouroriginal stellar rotation). As a first potential issue, the direct effect of assumptionthattheSNyieldspredominantlydependontheCO metallicityontheSNyieldsofnon-solarmodelsisnottakeninto coremassappearstobemorereliablefor26Althanfor60Fe. account.Indeed,ourgridofyieldstakesintoaccounttheeffectof AsecondpotentialissuestemsfromSNyieldscomingfrom metallicityonstellarevolutionprocesseslikethemasslossand non-rotatingmodels. Rotation has proven to be efficient at en- the late stagecore sizes. Butthe directinfluenceof metallicity, hancing the 26Al production during central H-burning, mostly that is the differing chemical composition of the stellar matter thankstoanincreasedconvectionandmixing.Ifrotationwereto fromwhichthenucleosynthesisproceeds,isnotconsidered.For modifytheyieldsoftheC/Neshell-burningepisodesinasimilar instance, we have shown before that an initial subsolar metal- manner, the corresponding26Al and 60Fe production would be licityleadstobiggerCOcore,hencepresumablytohigher60Fe higherthanthosecomputedbyLC06.Inaddition,increasingthe SN yields; however,this should be mitigated by the fact that a shellconvectionefficiencywouldalsoalterthefinalmass-radius lowermetallicityalsomeansreducedinitialquantitiesofFeand relation on which our estimate of the explosive nucleosynthe- CNOforthesynthesisof60Fe.Since60Feisapuresecondaryel- sis depends. We should also mention that, in the grid of yields ement,itislikelythatourgridoverestimatesthe60FeSNyields we are using, the ν-process5 has not been taken into account. of the subsolar metallicity models. Regarding 26Al, the direct impactofthemetallicityislessstraightforward.Indeed,mostof 5 Intheν-process,protonsliberatedbytheν-spallationofe.g.20Ne, the 26Al released by supernovae is of explosive origin and the 16O,23Na,and24Mgcaptureon25Mgtoform26Al. 10 PierrickMartinetal.:Predictedgamma-raylineemissionfromtheCygnuscomplex Thisproductionchannelwasestimatedtoenhancetheexplosive quantitativeanalysisoftheimpactofthenew60Felifetime,see 26Alyieldsby30-50%(Woosley&Weaver1995;Timmesetal. Vossetal.2009). 1995).Thesevaluesmayhavetoberevisedwithmoresophisti- We also emphasise that all the results presented so far are catedandrealisticneutrinospectraandtransport. basedona“classical”IMFslopeof1.35(Salpeter1955).Since More generally, the nucleosynthesis yields should be re- the universalityof thisparameteris still a matterof debate,we garded as sensitive to most ingredients of stellar models, such showinFig.6howthedecaylightcurvesforCygOB2varywith asconvectionandmixing,opacitiesornuclearreactionrates.In theslopeoftheIMF.Theimpactisfairlystraightforward:aflat- 3.2,wecomparedtheLC06andWH07setsofyieldsandshowed terIMFmeansmoreverymassivestars(foragiventotalnumber thatthechoiceofmasslossprescriptionortheimplementationof ofmassivestars)andthe1809keVpeakat3-4Myrattributable convectivephysicsdoessignificantlyaffectthe nucleosynthesis tothewindsandSNeofWRstarsisconsequentlystronger,while of26Aland60Fe.Wedidnotobservedrasticvariationsbymany asteeperIMFmeansfewerverymassivestars,whichreducesthe orders of magnitude in the outputs of the two grids of stellar 1809peakfluxat3-4Myrandenhancesthefluxat15-16Myr,as models,but the typicaldifferencesare high enoughto possibly aresultofmoreSNexplosionsof13-15M stars. ⊙ modify our conclusions. Therefore,although they were not in- InconnectionwiththepotentialeffectoftheIMFslope,we cludedinourpopulationsynthesis,wequalitativelyassessedthe want to draw attention to the possible impact of binarity. It is impactthattheWH07yieldswouldhaveonthepredicteddecay commonlyacceptedthatforstellar clustersmoredistantthan a fluxesfromtheCygnuscomplex. fewkpc,mostoftheobservedpoint-sourcesassumedtobemem- Forstarsinthe20-120M massrange(whichdominatethe bersoftheclustersmayactuallybeunresolvedbinariesormul- ⊙ current nucleosynthesis activity of the Cygnus complex), the tiplesystems.Thecensusoftheirstellarcontentisconsequently 26AlwindandSNyieldsoftheWH07modelsareclearlylower biased,particularlyforpurelyphotometricapproaches,buteven than those of MMP05 and LC06 (see Figs. 2 and 3). We esti- when relyingonspectroscopy.Kobulnicky&Fryer (2007) find matedthattheuseoftheseyieldsassuchinourpopulationsyn- thatthebinaryfractionofCygOB2isveryhigh,andmaywell thesiswouldlowerthe1809keVfluxbyatleastafactorof2and becloseto1;moreover,itseemsthatmassivestarspreferentially thusleadtoamismatchbetweenpredictionsandobservationsif havemassivecompanions.We thereforewonderaboutthepos- the Cygnus complexhas nearly solar metallicity. Nevertheless, sibilitythatasubstantialfractionofthe100-120OstarsofCyg the WH07 modelsdo notincludestellar rotation,whichturned OB2,manyofwhichhavebeenidentifiedphotometricallyfrom outtostronglyenhancethe26Alyields.Then,keepingthehope- the2MASSdata,mayinfactbeunresolvedmassivebinariesor fully morerealistic yieldsof MMP05for the 26Al wind contri- multiplesystems.Ifthisweretrue,thenumberofmassivestars bution, the lower 26Al SN yields of the WH07 models are not in our population synthesis would have to be increased, which expectedtoseriouslyaltertherisingpartoftheearly1809keV could lead to higher predicted fluxes. To properly account for lightcurvethatweobservetoday(seeFig.5),becausethelatter thisbinarityeffect,however,thevariousobservationsdevotedto is dominatedby thewind contribution.Thisisclearly apparent CygOB2(andtheotherOBassociations)havetobethoroughly inFig.6forthesimulationofCygOB2only:whenthefirststars compared to clearly identify which fraction of the objects as- of the coevalcluster explode,at ∼3.5Myr, the flux has already sumedtodayassinglemassivestarsmayactuallyhidemultiples, risento70-80%ofitsmaximumvalueandisalreadyconsistent butthisisbeyondthescopeofthepresentpaper. withtheobservedvalue.Consequently,eveniftheWH07mod- elsmayleadtoless26AlbeingejectedbySNe,thewindyields 4.2.Thespatialdiffusionof26Al ofthepresentlyavailablerotatingmodelsalonecanaccountfor mostoftheobserved26Aldecayflux.Regarding60Fe,theWH07 Intheprevioussections,wemadeuseofthephotometricaspects yields are not very different from the yields we have been us- oftheINTEGRAL/SPIobservationsofCygnusalone.Herewe ing (MMP05 models extrapolatedthrough CO core mass from wanttoextendourinterpretationoftheSPIresultstothespatial LC06)overthe20-120M massrange(seeFig.4).Wethusex- ⊙ andspectrometricinformationprovidedbythisinstrument. pectthe1173/1332keVlightcurvetostillbeconsistentwithour The numerous massive stars of the Cygnus complex are upperlimiteventiftheWH07yieldswereused. expected to have blown one or several superbubbles (SBs) WethereforebelievethattheWH07setofyieldswouldnot throughthe combinedaction of their winds andSN explosions modifyourresultsforthesolarmetallicitycaseifwestillusethe (Castoretal. 1975; Weaveretal. 1977; Tomisaka&Ikeuchi enhanced26Al wind yields of the rotating models. A definitive 1986; MacLow&McCray 1988; Chuetal. 2004; Chu 2008). conclusionwouldrequirefullyimplementingWH07inourpop- SoftX-rayobservationsbyHEAO-3andROSAThaverevealed ulationsynthesiscode(inorder,forinstance,totakethespecific a large structure in the Cygnus direction that may well be the SN yields to CO core mass relation of this grid properly into outermostpartofagiganticSB,thecentreofwhichisconcealed account). More generally, as alternative sets of yields become byinterveninginterstellarmaterial(Cashetal.1980).Fresh26Al available,especiallyfromrotatingmodels,itwillbeinteresting isexpectedto beinjectedinsidethehotandtenuousinteriorof to see if ourconclusionsstill hold.Evenin the absenceof ma- thisSB, andits diffusiononlargescaleswillthenbe drivenby jorimprovementsofthestellarmodels,theevolutionofsomeof thegrowthofthatstructure. theirinputsalone,suchasnucleardata,canalterthenucleosyn- Undercertain simplifications,a self-similaranalyticalsolu- thesis results. Some critical reaction rates, such as α(2α,γ)12C tioncanbefoundforthegrowthoftheSB(Castoretal.1975): or 12C(α,γ)16O, are known to have strong effects on the final yields(seeforinstanceTuretal.2009).Asanillustrationofthe R =66×L1/5n−1/5t3/5 pc (1) uncertaintieson nucleardata, the half-lifeof 60Fe was recently s 38 1 6 revised to 2.6Myr, instead of the 1.5Myr value we have been V =40×L1/5n−1/5t−2/5 kms−1 (2) s 38 1 6 usingthroughoutthiswork(Rugeletal.2009).Intermsofpre- dicted1173/1332keVlightcurvesfortheCygnuscomplex,how- where R and V are the outer radius and velocity of the SB, s s ever,thislongerlifetimesimplyimpliesanevenloweremission respectively,givenasfunctionsofthemechanicalluminosityof at early times, still consistent with our upper limit (for a more thecentralclusterL (in1038ergs−1),theuniformdensityofthe 38