Table Of Content

X-RAYS, PRE-MAIN SEQUENCE STARS AND PLANET FORMATION E.D. Feigelson Department of Astronomy & Astrophysics, PennsylvaniaState University,University Park PA 16802 USA Abstract 1990s established that late-type stars exhibit their high- est levels of magnetic activity during their PMS phases The study of magnetic activity of pre-main sequence (Feigelson& Montmerle 1999).But important puzzles re- (PMS) stars through their X-ray emission is entering a mainedunanswered.Domagneticfieldsinfullyconvective mature phase. We review here two recent developments. stars descending the Hayashi tracks arise from the same First, we present some early findings from the Chandra 5 OrionUltradeepProject(COUP)relatingtotheage-acti- dynamomechanismsasinsolar-typestars,wherethefields 0 aregeneratedintheshearlayerbetweenradiativeandcon- vity-rotation relations of PMS stars. COUP is a nearly 0 vectivezones?AretheX-rayflarescommonlyseeninPMS continuous exposure of the Orion Nebula Cluster over 13 2 stars generated in magnetic loops with both footprints in days and gives the most comprehensive view to date of n thestellarsurface,asinthe Sun,ormightthey arisefrom X-ray emission from the PMS population. We find that a reconnectionoffieldlines extending outto the protoplan- J the activity-rotation connection that dominates main sequence activitybehaviorisabsentinPMSstars,although etarydisk?Howdoesmagneticactivityevolvewithstellar 2 age?Dothehighenergyprocessesinvolvedinflaringhave 1 theactivity-agerelationispresentaswellaslinkstostellar a significant impact on the complex physics of the disk mass and volume. 1 Second, we discuss investigations of effects of PMS including accretion, outflows, and condensation into solid v bodies such as planets? X-rays on their cold protoplanetary disks. It is possible 3 The Chandra X-ray Observatory with its ACIS CCD that the the ionization, heating and spallogenic effects of 2 imaging array, a NASA Great Observatory launched in 2 high energy photons and particles produced in PMS stel- 1999, is particularly well-suited to PMS stellar studies. 1 larflareshavesignificanteffectsondiskdynamics,heating 0 and chemistry. X-ray ionization may induce MHD turbu- Its arcsecond imaging and subarcsecond astrometry per- 5 mitsunambiguousassociationsbetweenX-raysourcesand lence in the disk which, in turn, may regulate disk evolu- 0 stellar counterparts,evenin rich and crowdedyoung stel- tionandplanetformationprocessessuchasJovianplanet / h migration. lar clusters.Wide binaries and multiple systems are often p resolvedonscalesdowntotensofAU.Thelownoiseofits - CCDdetectorsandexcellentrejectionofcosmicrayevents o Key words:Stars:X-rays– OrionNebula – Planetforma- tr tion – macros: LATEX permits detection of sources with fluxes as low as 1 pho- s tonperday.Themirrorsurface,polishedtonearlysingle- a atom precision, efficiently reflects photons with energies v: up to 10 keV (wavelength 1 ˚A); photons above ≃ 3 keV Xi 1. Introduction can penetrate even the densest molecular cloud. Chan- dra’s sensitivity to magnetic activity rivals the sensitivity ar The surface activity of stars during their PMS phases are of8-meter-classtelescopestophotosphericemissioninthe not easily studied. The nearest active regions of star for- optical and near-infrared bands. The ACIS detector pro- mation are ≃ 150 pc distant and the stars often suffer duces a photon list in four-dimensions: right ascen sion, interstellarobscuration.Manytechniquesusefulforstudy- declination,photonenergy,andphotonarrivaltime.Spec- ing the stellar surfaces of closer and unobscured stars tralandtemporalanalysisprovidesaconsiderablebodyof – such as Zeeman spectroscopic measurements, Doppler information about the magnetic activity, the line-of-sight imaging, ultraviolet chromospheric indicators, radio con- obscuration, and occasionally about the circumstellar en- tinuumflaring–arerestrictedtotheclosestandbrightest vironment. PMS stars.However,elevated X-rayemission with strong flaring is ubiquitous from the early protostellar stages 2. The Chandra Orion Ultradeep Project through the T Tauri phases across the Initial Mass Func- tion, and can penetrate through hundreds of magnitudes TheOrionNebulahashistoricallybeenanextraordinarily of visual absorption. fruitfulsiteforthestudyofstarformationandearlystellar X-ray have thus become an important indicator for evolution. The Orion Nebula Cluster (ONC) has ≃ 2000 PMS surface activity, and studies during the 1980s and members concentrated in ≃ 10′ (1.3 pc) region with the Proc. 13th Cool StarsWorkshop, Hamburg,5–9 July 2004, F. Favata et al. eds. best characterized stellar Initial Mass Function ranging oped a model of rotational deceleration in main sequence from a ≃ 38 M⊙ O6-7 star to ∼ 3 MJ brown dwarfs. starsbasedonmagnetizedstellarwindsandWilson(1963) The OB members, known as the Trapezium, illuminate showed that stellar chromospheric activity was stronger Messier42,thenearestandbrightestHIIregioninthesky. in ZAMS stars like Pleiads than in older disk field stars. It is a blister HII region on the near side of Orion A, the Together, these studies led to an elaboration of empiri- nearest giant molecular cloud. Protoplanetary disks sil- cal links between age, rotation and activity such as Sku- houetted againstthe brightnebula havebeen directlyim- manich’s (1972) R ∝ v ∝ t−1/2 which provided a HK rot agedbytheHubbleSpaceTelescopearoundseveraldozen basisforextendingsolar-typedynamomodelstolate-type ONC members. An arcminute northwest of the Trapez- stars. ium and behind it in the molecular cloud lies OMC-1, the nearest massive molecular cloud core where star for- Inthemodernformulationofthisα−Ωdynamo(Schri- mation is now taking place. These include a dense clus- jver & Zwaan 2000), the magnetic field generation scales ter of intermediate-mass stars associated with the Orion with the rotational shear between the radiative and con- HotCoreandtheBecklin-NeugebauerObjectproducinga vective zones at the tachocline, which presumable scales powerful uncollimated molecular outflow, and protostars with the measurable surface rotational velocity. Talks at producing collimated Herbig-Haro outflows in the Orion laterCSSSworkshopsextendedthe activity-rotationrela- 1-South core. tionshiptopost-MSstarsintidally-lockedbinaries(Gunn The rich, crowded and obscured stellar field of the etal.1998)butfoundithasdifficultywhenappliedtofully OrionNebulaprovidesaperfectcomplementtoChandra’s convective M dwarfs (Valenti et al. 1998) and completely capabilities. Several instrument teams studied the region breaks down for L dwarfs (Mohanty & Basri 2003). It is inthe beginningofthe missionproducingaseriesofvalu- not clear that either very cool dwarfs or fully convective able studies (e.g. Schulz et al. 2001;Feigelson et al. 2002; PMS stars can exhibit the α−Ω dynamo which requires Flaccomio et al. 2003a). These groups then joined forces magnetic field generation at the tachoclinal interface be- and, togetherwith Orionexperts atother wavebandsand tween the radiation and convective zones. The possibility theoretical astrophysicists, proposed an extended obser- of convective zone dynamos was discussed as early as the vation known as the Chandra Orion Ultradeep Project 2nd CSSS workshop by Gilman (1981) and are being se- (COUP).Itconsistsofa9.7-dayACISexposuretakenover riously considered today for these stars. 13.2days,anearly-continuousobservationwithgapsonly during perigee of the satellite’s orbit around Earth. This givessuperbsensitivityoflogL =27.0erg/s(0.5−8keV Whilethetemporaldecayofmagneticactivityinmain band), sufficient to detect thexcontemporary active Sun, sequencelate-typestarsbetweenagesof∼0.1to∼1Gyr and uniquely long temporal coverage to study flaring for has long been well-established from chromospheric and 1616sources.NearlyallofthesearePMSstarsintheONC coronal studies of open clusters, and has recently been or OMC-1 regions. extended to the 1 − 10 Gyr epoch (Gu¨del et al. 1997; Feigelson et al. 2004). At the 7th CSSS workshop, Simon COUP has many scientific objectives including im- (1992) argued that stellar activity was physically depen- proving the census of ONC members, studying plasma dententirelyonrotationwithonlyanindirectrelationship propertiesinpowerfulstellarflares,elucidatingtheorigins with age.The situation for PMS stars has againbeen un- andevolutionofPMSX-rayemission,findingnewembed- clearwhetherthisargumentextendsintothePMSregime ded stars, studying the magnetic activity of subsamples where a monotonic rotational deceleration is replaced by such as OB stars, brown dwarfs, and stars with proplyds acomplicatedcompetitionbetweengravitationalcontrac- (i.e. (”PROtoPLanetarY Disks”) . We will also use the tion,internalviscouscoupling,andexternalmagneticcou- findings to improve theoretical studies of the effects of pling between the stellar surface and the protoplanetary high energy photons and particles on disk processes and disk (Barnes 2003). evolution. The reader is encouraged to examine the orig- inal papers, some appearing in a 2005 dedicated issue of the Astrophysical Journal Supplements, for detailed and Analysis of ROSAT observationsofyoungstellar clus- quantitative findings. terssuggestedthatmagneticactivitymeasuredintheunits logL /L , essentially the magnetic dissipation per unit x bol area on the stellar surface, increases during the first few 3. Age-activity-rotation relationships Myr to the saturation level logL /L ≃ −3 as the pro- x bol Here we restrict our discussion of COUP results on one toplanetarydisk disappears,anddeclines ontimescales of topic:extendingthe relationshipsbetweenstellaractivity, 107 −108 yr in a mass-dependent fashion (Flaccomio et ageandrotationintothePMSevolutionaryphases.These al. 2003b). However, uncertainties arise from possible in- relationships have been a foundation of our understand- completeness in the PMS samples,particularly for fainter ing of stellar magnetic activity, and have long been dis- weak-lined T Tauri stars and from unresolved multiple cussed at Cool Star Workshops. Schatzman (1962) devel- systems. do show strong associations: it scales with stellar mass and functions of stellar radius including surface area and internalvolume.Oneoftheserelations,plottingX-raylu- minosity against stellar volume, is shown in Figure 2. Similar results were obtained from earlier Chandra studiesoftheONCbutnoconsensusoninterpretationhas emerged (Feigelson et al. 2003, Flaccomio et al. 2003a, Stassun et al. 2004). One one hand, a standard α − Ω dynamo scenario canbe applied. The displacement of the Orion stars below the saturated level seen in MS stars in Figure1canbeexplainedbyadifferenceintheconvective zone depth and turnover time. When the Rossby number rather than rotational period is plotted, the Orion stars fall into the supersaturated regime where logL /L ≃ x bol −3.5 to −4.0 is typical. The dependence on mass may be unexplained but is also seen in a mass-dependence of the saturation level in main sequence stars (Pizzolato et al. 2003). Figure1. X-ray emissivity in units of logL /L plotted x bol against photometric rotational periods for Orion Nebula stars. The Orion stars are shown as black dots (time- averagedlevels includingflares),mainsequencedwarfs are shown as grey dots (Pizzolato et al. 2003, Messina et al. 2003), and the quiet-active Sun is shown as large circles. Orion X-ray luminosities are from the COUP study in the 0.5 − 8 keV band corrected for absorption (Getman et al. 2005), bolometric luminosities are from Hillenbrand (1997) and rotation periods are from Herbst et al. (2002). Preibisch et al., in preparation. 4. COUP results on age, activity and rotation The ONC probably provides the best possible sample for study of these questions. Cluster membership has been carefully determined into the brown dwarf regime, nearly a thousand stars have masses and ages estimated from placement on the Hayashitracks in the HR diagram,and several hundred members have surface rotational periods Figure2. X-ray luminosity logL (erg/s) plotted against x measuredfromphotometricmodulationofcoolstarspots. stellar volume (V⊙) for nearly 600 well-characterized The COUP findings on these issues are thus very promis- Orion stars. The plusses are quiescent levels and dots are ing avenues for understanding the origins of magnetic ac- time-averaged levels for COUP sources while arrows show tivity in fully convective stars. The work presented here X-ray nondetections. Preibisch et al., in preparation. in preliminary form will appear in COUP studies led by Thomas Preibisch based on the COUP source properties tabulated by Getman et al. (2005). On the other hand, one might emphasize the absence Figure 1 shows the X-ray emissivity logL /L as a ofaL −P relationandthe strengthoftheL −Mass− x bol x rot x function of rotational period for COUP Orion stars com- Volume relationships. The 5 orders of magnitude range pared to main sequence stars. Note that the COUP ob- in L is largely attributable to the relationships. Mag- x servation is so sensitive that there are virtually no X-ray netic activity dependencies on mass and volume in main nondetections in the sample. No obvious association be- sequence stars are known but are much weaker than seen tweenX-rayemissivityandrotationisseen.Averagedover in the PMS population. The importance of volume and the Initial Mass Function, T Tauri X-ray levels are typ- unimportance of rotation is qualitatively expected from ically slightly below the main sequence saturation level. somemodelsofdistributedturbulentdynamos,wherethe But other plots of X-ray emission against other variables magnetic field is both generated and transported in the convection zone. Indeed, it seems difficult to understand howasolar-typedynamocouldpossiblyoperateinafully convective star where no tachocline or differential rotational shear layer is expected. The difficulty in adjudicating this debate is the absence of quantitative predictions of dynamo models for fullyconvectivestars.Whilegeneraldependenciesbetween surfaceactivityandinteriorparameterscanbeestablished for solar-type α−Ω models (Montesinos et al. 2001), the modelsarenotstrictlypredictive.Thestateofdistributed dynamomodelsisevenmoreprimitivedespitesomerecent self-consistent calculations (e.g. Ku¨ker & Ru¨diger 1997, Kitchatinov & Ru¨diger 1999,Ku¨ker & Stix 2001).Efforts mustbemadetoestablishpredictedrelationshipsbetween magnetic field generationandthe principalglobalparam- eters–rotation,massandvolume–fordifferentclassesof dynamo mechanisms. Simon(1992)arguedthattheactivity-agerelationwas not physically fundamental in main sequence stars, but ratheraderivativeeffectfromactivity-rotationandrotation- age relations. In contrast, we find in COUP studies an activity-age relation in PMS stars even in the absence of an activity-rotation relation. Figure 3 shows the evolution of both the X-ray luminosity logL and emmisivity x logL /L for stellar ages ranging from logt≃6.0 to 9.5 x bol yr. Again we emphasize that no statistical biases should be present in the COUP study: the sample is complete in both in its stellar sample and in the X-ray detections, and large enough that mass-stratified subsamples can be independently examined. Taken together, the COUP results in Figures 1 − 3 arequitedifferentfromthoseseeninmainsequencestars. One possible explanation is that the magnetic dynamo mechanism,whateveritmaybe,issaturatedinthestellar volume or surface. This would account for a drop in surface activity as the star descends the convective Hayashi Figure3. Median X-ray luminosity L (top) and emissiv- tracks that is uncorrelated with rotation. Alternative ex- x ityL /L as a function of stellar age for lightly absorbed planationswouldbebasedonchangesininteriorstructure, x bol stars in the COUP study of the ONC in comparison with such as rotational coupling between surface and interior mainsequencestarsofvariousages.ThePleiades,Hyades layers or the growthof a radiative core in older and more and solar neighborhood field star results are obtained from massive PMS stars. ROSAT studies. The solid dots show the 0.9−1.2M⊙ (G-type on the main sequence) stars, squares show the 5. Magnetic flares and the protoplanetary disk 0.5−0.9M⊙ (K-type on the main sequence) stars, and crosses the 0.1−0.5M⊙ (M-type on the main sequence) The circumstellar disks around PMS stars where plane- stars. Preibisch et al., in preparation. tary systems form are generally considered to consist of cool, neutral molecular material in thermodynamic equilibrium.Withcharacteristictemperaturesrangingfrom20 to1500K,theyemitprimarilyintheinfraredandsubmil- within the context of a solar nebula in thermodynamic limeter bands. Except for stars residing close to O stars equilibrium.Theseincludetheenormousquantitiesofflash whose ultraviolet radiation can destroy disks by photoe- melted chondrules, the calcium-aluminum-rich inclusions vaporation (Hollenbach et al. 2000), protoplanetary disks (CAIs) with strange combinations of short-lived radioiso- have, until the recent studies outlined here, been treated topic anomalies, and the gas-rich grains with spallogenic as closed and isolated structures. 21Neexcessescorrelatedwithenergeticparticletrackden- However,therearelong-standingcharacteristicsofan- sities (see reviews by Jones et al. 2000, Goswami & Van- cient meteorites which are exceedingly difficult to explain hala 2000, and Woolum & Hohenberg 1993). These findings still are deeply disturbing as no single, straightfor- 6. Disk ionization sources wardastrophysicalexplanationcanaccountforallofthem. The first discussion of protoplanetary disk ionization by Some problems seem to be most readily explained by the Gammie (1996) cited Galactic cosmic rays as the likely injectionofrecentlysynthesizedradionuclidesfromsuper- penetrating agent. In standard structure models, ioniza- novae or other stellar ejecta into the solar nebula, while tion would penetrate all the way to the midplane in the others seem to require spallationby MeV particles within outer regions of the disk while a midplane neutral ‘dead thenebula.Otherphenomenaareoftenattributedtoheat- zone’wouldprevailinthe innerdisk.Oneconcernis whe- ing by shocks from supersonic bodies in the nebula. therthemorenumerouslowenergycosmicrayscanpene- tratethetangledmagneticfieldsofthesurroundingmolec- ular cloud and the flow of partially ionized outflows from See full version at the young stellar system (analogous to the solar Forbush http://www.astro.psu.edu/users/edf/CS13.pdf effect). Desch et al. (2004) considers these problems and find that cosmic ray penetration is likely quite efficient. Figure4. Cartoon illustrating the geometry of magnetic An analogous treatment of disk ionization using PMS fields and high energy photons and particles with respect X-rays was developed by Glassgoldet al. (1997) and Igea to the protoplanetary disk. (Feigelson 2003) & Glassgold (1999) who showed that Compton processes increaseX-raypenetrationthroughcolumndensitiesofor- der1024−1025cm−3 (seereviewbyGlassgoldetal.2000). As cosmic ray fluxes should be roughly spatially uniform X-ray astronomical studies of PMS stars have a pow- butPMSX-raysfluxesincreaseasr−2orfaster(duetoab- erful role in developing models for these meteoritic mys- sorptionofthe softer photons)asone approachesa young teries. They show that magnetic reconnection flares, or- star, X-ray ionization will always exceed cosmic ray ion- ders of magnitude more powerful and frequent than the izationsufficiently close to a star.The cross-overdistance strongest flares seen in the contemporary Sun, are ubiq- for these two ionization sources is model-dependent but, uitous in solar mass stars throughout the PMS phases forPMSsolarlevelsaroundlogL ≃30erg/sandcharac- of evolution. After Einstein Observatory studies of PMS x tericplasmatemperaturesofkT ≃1−5keV,likelyoccurs X-rays, I suggested a link between X-ray flares and me- around a few AU (Matsumura & Pudritz 2003). teoritic issues (Feigelson 1982) and argued that ‘invoking Collisionalionizationofatomswithlowfirstionization high levels of energetic particle or radiation fluxes associ- potentials, such as potassium and sodium, may also be ated with magnetic activity in the early Sun is no longer important when disk temperatures exceed ≃ 800 K. Dif- ad hoc’ (Feigelsonet al. 1991).After Chandra studies like feringtreatmentsofthesemetalionsleadstoconsiderable COUP,thisstatementcanbestrengthened:Consequences differences in the inferred steady-state ionization level of of high energy radiation, both photons and particles, asso- the disk, even for constant external ionization rates (Fro- ciated with frequent and powerful magnetic reconnection mangetal.2002).Thetheoryofthe X-rayionizationrate events within PMS systems cannot be avoided. The main appears satisfactory,but calculations of the X-ray ioniza- uncertainty in applying this principle is the geometrical tion fraction depend on poorly established recombination relationshipbetweenthe flaringfieldlines anddisk region rates. of interest (Feigelson et al. 2002). Figure 4 provides a visual framework for these issues with a reasonable guess X-rayionizationeffectsareonlyoneofseveralcontrib- concerning the geometry. utorstotheverycomplexandnonlinearinterplaybetween the thermodynamics, dynamics (including magnetic field The question is no longer whether high energy pho- effects),gas-phasechemistryandgas-graininteractionsin tons and particles, far out of thermodynamic equilibrium protoplanetarydisks.Astronomicalobservationsofmolec- withthethermalmaterial,aregeneratedinPMSsystems. ular species are rapidly emerging, but suffer from insuffi- The questions are rather: What are the geometrical con- cientspatialresolutionandsensitivityandoftengivecon- figurations of the reconnecting fields? and What are the fusing results. Given the complexities of disk thermody- consequent effects on disk gases and solids? Feigelson et namics and chemical evolution, confident understanding al. (2002) gives a Chandra-based estimate of the high of X-ray chemical effects may await the ALMA interfer- energy particle irradiation of the disk and discusses the ometer which will image many molecules across the disks uncertain magnetic geometry. Implications for the mete- of nearby PMS stars. oritic anomalies of shortlived radionuclides are discussed by Leya et al. (2003), Gounelle et al. (2004), Marhas & Goswami(2004),Deschetal.(2004)andreferencestherein. 7. Disk turbulence and its implications Here,Ibrieflyreviewrecentinvestigationsofthepossi- Provided a weak magnetic field was entrained in the disk ble effectsofX-ray(ratherthanMeVparticle)irradiation during collapse, the Keplerian shear in the ionized zone of the disk. interacts with the field to induce the magnetorotational instability, also known as the Balbus-Hawley instability, ItisthuspossiblethatX-rayemissionplaysanimpor- which quickly develops into a full spectrum of MHD tur- tantroleinregulatingthestructureanddynamicsofplan- bulence including both verticalandradialmixing.Ioniza- etarysystems.WeakerPMSX-raysourceswouldproduce tionfractionsaslowas10−11 maygiveadequatecoupling disks with weaker turbulence, stronger Type I migration which is achievedwhen the collisionfrequency ofneutrals withrapidinspiralling,resultinginplanetarysystemswith with ions exceeds the local epicyclic frequency (Blaes & ‘hot Jupiters’ like 51 Peg. Stronger PMS X-ray sources Balbus1994).Recentstudieshaveinvestigatedthegrowth might produce disks with stronger turbulence, random oftheinstabilityincludingeffectsofverticalstratification, walk radial motion so that Jovian protoplanets remain in finite conductivity,ohmic dissipation,ambipolardiffusion themid-diskregions;the resultingplanetarysystemsmay and Hall effects (Salmeron & Wardle 2003;Desch 2004). resemble those aroundυ And andour Sun. Randomwalk Thisionization-inducedturbulencemaybeofconsider- migrationweakensdiskgapsandspeedsupprotoplanetary able importance for planet formation and dynamical evo- accretion.Thismayaccountforthemuchhighermassesof lution in several ways: active zone Jovian planets compared to dead zone terres- trialplanets(Matsumura&Pudritz2003).Jovianplanets 1. The radial viscosity associated with the active turbu- undergoingturbulent randomwalks alsoaccrete their full lentzonemaycausetheflowofmaterialfromtheouter masses more rapidly than those in non-turbulent disks, disk into the inner disk, and thereby into the bipolar which can alleviate planet formation timescale problems outflowsandontotheprotostar.Thismaysolvealong- in some earlier models (Rice & Armitage 2003). standingprobleminyoungstellarstudies:acompletely Finally, we note that PMS X-raysare a major sources neutral disk should have negligible viscosity and thus of ionization at the base on outflows from protostellar cannotefficiently be anaccretiondisk.Armitageetal. disks which produce the emission line Herbig-Haro ob- (2001)suggestthatalong-termdiskinstabilitymaybe jects and molecular bipolar outflows (Shang et al. 2002, also occur: when the dead zone temperature exceeds Ferro-Fontań et al. 2003). This is a profound result: if ≃ 800 K so that collisional ionization rises, the entire low-massPMSstarswerenotmagneticallyactiveandpro- innerdiskmaysuddenlybecomeviscousandspiralto- fusely emitting penetrating photoionizing radiation, then wards the star. Such bursts of accretion may give rise the coupling betweenthe Keplerianorbits inthe disk and to the FU Orionis phenomena and episodic accretion the magnetocentrifugally orbits spiralling outward per- seen as symmetric shocks in large-scale Herbig-Haro pendicular to the disks might be much less efficient than outflows. we see. Without powerful disk outflows, molecular cloud 2. Ionization-inducedturbulence willproduce density in- energetics,diskevolutionandevenplanetformationwould homogeneities which can overwhelm other extensively be considerably altered. studied dynamical structures – such as spiral instabilities, gaps and streams – that emerge from dynamical 8. Other X-ray effects on disks interactions between protoplanets and gaseous disks. A possible crucial consequence may be that turbulent It has been known for 20 years that X-ray irradiation of density perturbations will affect gap formation and a molecular environment will induce a complex series of produce stochastic radial torques on the protoplanet. molecular-ion and radical chemical reactions. This The protoplanet’sradialmotionmaythenresemblesa was first applied to protoplanetary disk chemistry by Ai- random walk rather than the rapid inspiralling called kawa & Herbst (1999) and further studied by Aikawa & Type I migration. Several groups have made MHD Herbst (2001) Markwick et al. (2002) and Semenov et al. studiesofthis process(Nelson&Papaloizou2004;Pa- (2004). The latter authors find that, from the chemical paloizou & Nelson 2003; Winters et al. 2003; Menou perspective, X-ray ionization is a critical ionizing source & Goodman 2004;Laughlin et al. 2004;Hersant et al. in the surface disk layers close to the star (r < 10 AU) 2004). Results from one of these calculations is illus- andalsoatdeeperintermediatelayersthroughoutthedisk trated in Figure 5. where it drives a rich ion-molecular chemistry (Figure 6). CN, HCO+ and C2H abundances may be good tracers of photoionization effects, though it is often difficult to See full version at distinguish X-ray and ultraviolet irradiation from global http://www.astro.psu.edu/users/edf/CS13.pdf disk observations. The CN/HCN ratio is elevated in the disks around X-ray luminous T Tauri and Herbig Ae/Be Figure5. Three-dimensional magnetohydrodynamical sim- stars pointing to photochemistry, but again the roles of ulation of a disk with a protoplanet. The top panel shows UV and X-ray effects are again intertwined (Thi et al. a neutral disk with spiral instabilities and streams which 2004). X-ray heating should also lead to ice evaporation cause Type I migration. The bottom panel shows the same and enhanced gaseous abundances of molec ules such as model with turbulent inhomogeneities caused by instabili- methanol,thoughthishasnotyetbeenseen.X-rayioniza- ties arising from ionization. (Laughlin et al. 2004) tion may also be manifested in the chemistry of infalling envelopes around Class 0 protostars (Doty et al. 2004). When PMS X-rays are absorbed by dust grains, However, evidence for X-ray ionization-induced chemical the localized heating may desorb molecules in the grain speciesisoftenmixed;forinstance,Ceccarellietal.(2004) mantle and increasing some molecular abundances in the find H2D+ abundance in the TW Hya disk is consistent gas phase (Najita et al. 2001). Conceivably, such spot with cosmic ray ionization alone. heating may produce liquid water which would change mineral chemistry; this is a suggested explanation for an X-rayabsorptioncontributetothewarmingofouter unexpected infrared emission band attributed to calcite molecular layers of the disk. Glassgold et al. (2004) (Ceccarelli et al. 2002). show that, in the outermost layer, the gas is heated to Another X-ray heating effect in the disk may be the 5000 K, far above the equilibrium dust temperature. One flashmeltingofdustballsproducingmeteoriticchon- consequence is a thick layer of warm CO responsible for drules and/or CAIs (Shu et al. 2001).Such models are the strong CO infrared bands seen from several young attractivein their integratedtreatmentofspallogeniciso- disks. However, the relative importance of heating from topicanomaliesandflashmelting.Buttheyarechallenged X-ray irradiation and mechanical heating from external by the need to transport the mm-cm-sized melted rocks wind-disk interaction or shock dissipation of internal tur- fromthevicinityoftheflaresnearthestartotheAsteroid bulence is unclear. Gorti & Hollenbach (2004) model the Belt where they are now prevalent. heating to partially dissipated older T Tauri disks. X-ray heating is important in the outer disk layers and the disk edge assumed to lie 0.1 AU from the star. They also pre- 9. Final comments dict a correlation between the neutral oxygen 63µm line strength and X-ray irradiation. X-ray excitation of disk The COUP study is solidifying our knowledge of the phe- molecules is a plausible origin for the ro-vibrational lines nomenology of X-ray emission and magnetic activity in ofH2 seeninseveralTTauristarsincluding,surprisingly, PMS stars. It confirms earlier Chandra findings that the oneclassifiedasadisk-freeweak-linedTTauristar(Wein- age-activity-rotation relationships seen in main sequence traub et al. 2000, Bary et al. 2003). late-type stars do not apply in the earlier phases of evolution. The main predictors of stellar activity is a weave of stellar mass-luminosity-volume variables, not rotation. However,ourunderstandingofthisfindingisnotreaching 300 clarity.We do not know whether to treat the effect as ev- 200 m ediatelayer isnduiespmnecres(aif.teou.r,raaatvdeadisrtirareingbtiumoteefd)stotarunrfdboaurrladendαtiffc−oenrΩevnedtcytdnivyaenmadomysno(aimm.e.eo,c)th.hae- AU Inter Thisisonlyoneofmanyavenuesofresearchbeingcon- Z, yer ductedwithCOUP,andCOUPisonlyoneofmanyefforts a l 100 ace in studying PMS stars using Chandra. From a purely as- urf Midplane tronomical viewpoint, Chandra will emerge with catalogs S oftensofthousandsofPMSstars,mostpreviouslyunstud- ied,inmanystarformationregionsatdistancesof0.5−10 0 0 100 200 300 400 kpc in the Galactic disk. This alone can give tremendous R,AU impetus to the study of PMS stars in the coming years. Observationalconstraintsonthe various aspects of X- ray irradiationof protoplanetarydisks are rapidly emerg- 10-3 ing.Impressiveadvancesininfraredinstrumentationhave beenachievedwiththeSpitzerSpaceTelescopeandground- 10-6 U U huj based detectors such as Visir on the ESO Very Large Xe10-9 1 AU 3 AU 10 AU 30 AU 100 AU 300 A 370 A TinesliegshctospewsilalnadwaTit-RtehCeShiognhtrheesoGluetmioinniaTndelessecnospitei.viStoymoef ALMA in the submillimeter band. 10-12 On the X-ray front, COUP is already providing two 10-105.01 0.1 1 10 100 lines of evidence that disk irradiation by X-rays is com- Z, AU mon. First, COUP detects 2/3 of Orion stars with disks imaged with the Hubble Space Telescope (J. Kastner et Figure6.Diskstructure(top) andionization fraction (bot- al., in preparation). In a few cases, X-ray absorption will tom) resulting from a protoplanetary disk model with a measuretheline-of-sightcolumndensitythroughthedisk. complex ion-molecular chemical reaction network. The Second,in a few of the brightestX-ray sourcesassociated solid line defining the top of the outer layer denotes the with very young PMS stars, we detect the 6.4 keV flu- 10−4 ionization level. (Semenov et al. 2004) orescent emission line (M. Tsujimoto et al., in preparation). We argue this arises from reflection off of the cold Gounelle, M., Shu, F. H., Shang, H., et al. 2004, Lunar Plan disk. The next generation of X-ray telescopes, unfortu- Inst Conf, 35, 1829 nately overadecadeinthe future,conceivablycouldinfer Gu¨del,M.,Guinan,E.F.,&Skinner,S.L.1997,ApJ,483,947 the structure of the disks through reverberationmapping Gunn,A.G.,Mitrou,C.K.,&Doyle,J.G.1998,inCoolStars, Stellar Systems, and the Sun, 10, 1257 of X-ray flares. Herbst W., Bailer-Jones, C. A., Mundt, R., Meisenheimer, K. Acknowledgements & Wackermann,R. 2002, AA, 396, 513 This work was supported by Chandra Guest Observer grants Hersant, F., Dubrulle,B. & Hure,J. M. 2004, AA,in press GO3-4009A supporting COUP and GO4-5505X supporting Hollenbach, D. J., Yorke, H. W., & Johnstone, D. 2000, in studyofolderdisks.ThomasPreibisch(MPIfR)kindlyshared Protostars and Planets IV, 401 COUP results prior to publication. Dmitry Semenov (MPIA), Jones, R. H., Lee, T., Connolly, H. C., Love, S. G., & Shang, Gregory Laughlin (UC Santa Cruz) and Al Glassgold (UC H.2000, in Protostars and Planets IV,927 Berkeley) provided valuable assistance on themanuscript. Kitchatinov,L. L. & Ru¨diger, G. 1999, AA,344, 911 Ku¨ker, M. & Ru¨diger, G. 1997, AA,328, 253 Ku¨ker, M. & Stix, M. 2001, AA, 366, 668 References Leya, I., Halliday, A.N., & Wieler, R. 2003, ApJ, 594, 605 Aikawa, Y.& Herbst, E. 1999, AA,351, 233 Laughlin, G., Steinacker, A., & Adams, F. C. 2004, ApJ, 608, Aikawa, Y.& Herbst, E. 2001, AA,371, 1107 489 Armitage,P.J.,Livio,M.,&Pringle,J.E.2001,MNRAS,324, Marhas, K. K. & Goswami, J. N. 2004, New Astro Rev, 48, 705 139 Barnes, S. A.2003, ApJ, 586, 464 Markwick,A.J.,Ilgner,M.,Millar,T.J.,&Henning,T.2002, Bary,J.S.,Weintraub,D.A.,&Kastner,J.H.2003,ApJ,586, AA,385, 632 1136 Matsumura, S. & Pudritz, R.E. 2003, ApJ, 598, 645 Blaes, O. M. & Balbus, S. A.1994, ApJ,421, 163 Menou, K. & Goodman, J. 2004, ApJ,606, 520 Ceccarelli, C., Caux, E., Tielens, A. G. G. M., Kemper, F., Mohanty, S. & Basri, G. 2003, in Cool Stars, Stellar Systems, Waters, L. B. F. M., & Phillips, T. 2002, AA,395, L29 and the Sun, 12, 683 Ceccarelli, C., Dominik, C., Lefloch, B., Caselli, P., & Caux, Montesinos, B., Thomas, J. H., Ventura, P., & Mazzitelli, I. E. 2004, ApJL, 607, L51 2001, MNRAS,326, 877 Desch,S.J.,Connolly,H.C.,&Srinivasan,G.2004,ApJ,602, Najita, J., Bergin, E. A., & Ullom, J. N.2001, ApJ, 561, 880 528 Nelson, R. P.& Papaloizou, J. C. B. 2004, MNRAS,350, 849 Desch, S. J. 2004, ApJ,608, 509 Papaloizou, J. C. B. & Nelson, R. P. 2003, MNRAS,339, 983 Doty, S. D., Schöier, F. L., & van Dishoeck, E. F. 2004, AA, Pizzolato,N.,Maggio,A.,Micela,G.,Sciortino,S.,&Ventura, 418, 1021 P. 2003, AA,397, 147 Feigelson, E. D. 1982, Icarus, 51, 155 Rice, W. K.M. & Armitage, P.J. 2003, ApJL, 598, L55 Feigelson, E.D.,Giampapa,M.S.,&Vrba,F.J.1991, inThe Salmeron, R. & Wardle, M. 2003, MNRAS,345, 992 Sun in Time,658 Schatzman,E. 1962, Ann d’Astrophys,25, 18 Feigelson, E. D. & Montmerle, T. 1999, ARAA,37, 363 Schrijver,C.J. &Zwaan, C.2000, Solar and Stellar Magnetic Feigelson, E. D., Broos, P., Gaffney, J. A., et al. 2002, ApJ, Activity, NY:Cambridge University Press 574, 258 Schulz, N. S., Canizares, C., Huenemoerder, D., et al. 2001, Feigelson, E. 2003, in Stars as Suns: Activity, Evolution and ApJ, 549, 441 Planets, IAUSymp 219 Semenov,D., Wiebe, D., & Henning,T. 2004, AA,417, 93 Feigelson, E. D., Gaffney, J. A., Garmire, G., Hillenbrand, Shang, H., Glassgold, A. E., Shu, F. H., & Lizano, S. 2002, L. A.,& Townsley, L. 2003, ApJ,584, 911 ApJ, 564, 853 Feigelson, E.D.,Hornschemeier,A.E.,Micela, G.etal. 2004, Shu,F. H., Shang, H., Gounelle, M., Glassgold, A. E., & Lee, ApJ,in press T. 2001, ApJ,548, 1029 Ferro-Fontán,C.&deCastro,A.I.G.2003,MNRAS,342,427 Simon, T. 1992, in Cool Stars, Stellar Systems, and the Sun, Flaccomio,E.,Damiani,F.,Micela,G.,etal.2003a,ApJ,582, 7, 3 398 Skumanich,A.1972, ApJ,171, 565 Flaccomio, E.,Micela,G.,&Sciortino,S.2003b,As&Ap,402, Stassun,K.G.,Ardila,D.R.,Barsony,M.,Basri,G.,&Math- 277 ieu, R.D. 2004, AJ, 127, 3537 Gammie, C. F. 1996, ApJ,457, 355 Thi, W.-F., van Zadelhoff, G.-J. & van Dishoeck, E.F˜. 2004, Getman,K.V.,Flaccomio, E.,Broos,P.S.,etal.2005, ApJS, AA,in press in press Valenti, J. A., Johns-Krull, C. M., & Piskunov, N. 1998, in Gilman, P. A. 1981, in Cool Stars, Stellar Systems, and the Cool Stars, Stellar Systems, and the Sun, 10, 1357 Sun, 2, 165 Weintraub,D.A.,Kastner,J.H.,&Bary,J.S.2000,ApJ,541, Glassgold, A. E., Feigelson, E. D., & Montmerle, T. 2000, in 767 Protostars and Planets IV,42 Wilson, O.C. 1963, ApJ, 138, 832 Glassgold, A.E., Najita, J. & Igea, J. 2004, ApJ, in press Winters,W.F.,Balbus,S.A.,&Hawley,J.F.2003,ApJ,589, Gorti, U. & Hollenbach, D. J. 2004, ApJ,in press 543 Goswami, J. N. & Vanhala, H. A. T. 2000, in Protostars and Woolum,D.S.&Hohenberg,C.1993,inProtostars andPlan- Planets IV, 963 ets III,903