New Ideas in the Theory of Extrasolar Giant Planet

arXiv:astro-ph/9809084v18Sep1998NEWIDEASINTHETHEORYOFEXTRASOLARGIANTPLANETSANDBROWNDWARFSADAMBURROWS,BILLHUBBARD,andJONATHANLUNINEUniversityofArizonaMARKMARLEYNewMexicoStateUniversityandDIDIERSAUMONVanderbiltUniversityI.INTRODUCTIONThestudyofextrasolargiantplanets(EGPs*)andbrowndwarfsviareflexstellarmotion,broad–bandphotometry,andspectroscopyhasfinallycomeintoitsown.Dopplerspectroscopyalonehasrevealedabout15objectsinthegiantplanet/browndwarfregime,includingcompanionstoτBoo,51Peg,υAnd,55Cnc,ρCrB,70Vir,16CygB,and47UMa(Noyesetal.1997;Butleretal.1997;Cochranetal.1997;Marcy&Butler1996;Butler&Marcy1996;Mayor&Queloz1995;Lathametal.1989).ThedirectdetectionofGl229B(Oppenheimeretal.1995;Naka-jimaetal.1995;Matthewsetal.1996;Geballeetal.1996;Marleyetal.1996;Allardetal.1996;Tsujietal.1996)wasamilestonesinceGl229Bdisplaysmethanespectralfeaturesandlowsurfacefluxesthatareuniquetoobjectswitheffectivetemperatures(inthiscase,Teff∼950K)belowthoseatthesolar–metallicitymainsequenceedge(∼1750K,Burrowsetal.1993).Inaddition,thealmostcompleteabsenceofspectralsignaturesofmetaloxidesandhydrides(suchasTiO,VO,FeH,andCaH)isinkeepingwiththeoreticalpredictionsthatthesespeciesaredepletedintheatmospheresofallbuttheyoungest(hence,hottest)substellarobjectsandaresequesteredincondensedformbe-lowthephotosphere(Lunineetal.1989;Marleyetal.1996).Thewiderangeinmassandperiod,aswellastheproximityofmanyofthe*WeusethisshorthandforExtrasolarGiantPlanet,buttheterms“ex-oplanet”or“super–jupiter”areequallyappropriate.[1]2planets/browndwarfstotheirprimaries,wasnotanticipatedbymostplanetaryscientists.ThoughthetechniqueofDopplerspectroscopyusedtofindthesecompanionsselectsformassive,nearbyobjects,theirvarietyandexistenceisachallengetoconventionaltheory.Sincedi-rectdetectionisnowfeasible,andhasbeendemonstratedbytherecentacquisitionofGl229B,itiscrucialforthefutureofextrasolargiantplanetsearchesthatthespectra,colors,evolution,andphysicalstruc-tureofobjectsfromSaturn’smass(0.3MJ)to70MJbetheoreticallyinvestigated.II.EARLYCALCULATIONSOFTHEEVOLUTIONANDSTRUCTUREOFEXTRASOLARGIANTPLANETSA.GrayModelsEGPsradiateintheopticalbyreflectionandintheinfraredbythethermalemissionofbothabsorbedstellarlightandtheplanet’sowninternalenergy.InBurrowsetal.(1995)andSaumonetal.(1996),theEGPswereassumedtobefullyconvectiveatalltimes.Weincludedtheeffectsof“insolation”byacentralstarofmassM∗andconsideredsemi-majoraxes(a)between2.5A.U.and20A.U.Giantplanetsmayformpreferentiallynear5A.U.(Boss1995),butasthenewdatadramaticallyaffirm,abroadrangeofsemi–majoraxescannotbeexcluded.Inthesecalculations,weassumedthattheBondalbedoofanEGPisthatofJupiter(∼0.35).FortheBurrowsetal.(1995)study,weevolvedEGPswithmassesfrom0.3MJ(themassofSaturn)through15MJ.Whethera15MJobjectisaplanetorabrowndwarfislargelyasemanticissue,thoughonemightdistinguishgasgiantsandbrowndwarfsbytheirmodeofformation(e.g.,inadiskor“directly”).If51Pegbisagasgiant,itsradiusisonly1.2RJanditsluminosityisabout3.5×10−5L⊙.Thisbolometricluminosityismorethan1.5×104timesthepresentluminosityofJupiterandonlyafactoroftwobelowthatattheedgeofthemainsequence.Theradiativeregionencompassestheouter0.03%inmass,and3.5%inradius.ThestudybyGuillotetal.(1996)demonstratedthat51PegbiswellwithinitsRochelobeandisnotexperiencingsignificantphotoevaporation.Itsdeeppotentialwellensuresthat,evensoclosetoitsparent,51Pegbisstable.If51Pegbwereformedbeyond1A.U.andmovedinwardonatimescalegreaterthan∼108years,itwouldcloselyfollowaRp∼RJtrajectorytoitsequilibriumposition.B.Non–GrayModelsHowever,tocrediblyestimatetheinfraredbandfluxesandimproveupontheblackbodyassumptionmadeinBurrowsetal.(1995)and3Saumonetal.(1996),wehaverecentlyperformednon-graysim-ulationsatsolar–metallicityoftheevolution,spectra,andcolorsofisolatedEGP/browndwarfsdowntoTeffsof100K(Burrowsetal.1997).Figure1portraystheluminosityversustimeforobjectsfromSaturn’smass(0.3MJ)to0.2M⊙forthismodelsuite.Theearlyplateauxbetween106yearsand108yearsareduetodeuteriumburn-ing,wheretheinitialdeuteriummassfractionwastakentobe2×10−5.Deuteriumburningoccursearlier,isquicker,andisathigherluminos-ityforthemoremassivemodels,butcantakeaslongas108yearsfora15MJobject.Themassbelowwhichlessthan50%ofthe“primordial”deuteriumisburntis∼13MJ(Burrowsetal.1995).Onthisfigure,wehavearbitrarilyclassedas“planets”thoseobjectsthatdonotburndeuteriumandas“browndwarfs”thosethatdoburndeuterium,butnotlighthydrogen.Whilethisdistinctionisphysicallymotivated,wedonotadvocateabandoningthedefinitionbasedonorigin.Neverthe-less,theseparationintoMdwarfs,“browndwarfs”,andgiant“planets”isusefulforparsingbyeyetheinformationinthefigure.InFig.1,thebumpsbetween10−4L⊙and10−3L⊙andbetween108and109years,seenonthecoolingcurvesofobjectsfrom0.03M⊙to0.08M⊙,areduetosilicateandirongrainformation.Theseeffects,firstpointedoutbyLunineetal.(1989),occurforTeffsbetween2500Kand1300K.Thepresenceofgrainsaffectstheprecisemassandluminosityattheedgeofthemainsequence.Sincegrainandcloudmodelsareproblematic,therestillremainsmuchtolearnconcerningtheirroleandhowtomodelthem(Lunineetal.1989;Allardetal.1997;Tsujietal.1996).III.NEWIN