Star Formation, Brown Dwarfs, and Exoplanets This is a big topic Star formation is discussed in your text (a little), in the text by LeBlanc (a bit more), and in some review articles (linked on the course webpage)
ISM Summary
The ISM and Dense Cool Clouds Interstellar medium: The matter (gas and dust) that exists between the stars in a galaxy Mostly H, He, but also 1-2% grains that may be ice, silicates, or metals (in the physicist sense). Typical density ~10-21 kg m-3. Density of stars ~ 0.08 pc-3 in the solar neighborhood, or 3 X 10-24 g cm-3 assuming the stars are 0.5M. Number density of n ~ 1 cm-3 Overall, mass of ISM similar to that of stars. Small fraction of ISM is in molecular clouds, cold regions typically composed of molecular hydrogen, H2. These clouds are likely places for new stars to form.
Interstellar Gas Dust is ~ 1% of the ISM mass. ISM is mostly gas (primarily H) This can be neutral (H I), ionized (H II), or molecular (H2) Most of the H is neutral. At low ISM T, electrons are mostly in the ground state. No emission lines Absorption lines require the presence of UV photons
ISM Make-up Interstellar clouds Warm/Hot rarefied gas T ~ 1000 10000 K, n ~ 0.1 cm-3 (pressure equilibrium w/ clouds) Fills most of the ISM Hot bubbles T ~ 100 K, n ~ 1 10 cm-3, R ~ 10 pc About 5% of ISM Million K bubbles evacuated by SNe explosions Molecular clouds Very dense (characterized by large extinction) Neutron H combines to H2 21 cm emission plateaus n ~ 103 105 cm-3
All-sky map of neutral hydrogen. J. Dickey (UMn), F. Lockman (NRAO), SkyView http://antwrp.gsfc.nasa.gov/apod/ap010113.html http://antwrp.gsfc.nasa.gov/apod/ap970430.html Molecular (CO) map T. Dame (CfA, Harvard) et al., Columbia 1.2-m Radio Telescopes
Local Bubble Linda Huff (American Scientist), Priscilla Frisch (U. Chicago) http://antwrp.gsfc.nasa.gov/apod/ap020210.html Linda Huff (American Scientist), Priscilla Frisch (U. Chicago) http://antwrp.gsfc.nasa.gov/apod/ap020217.html A model of the local ISM. The Sun is believed to moving through the local interstellar cloud just on the edge of the local bubble.
Giant Molecular Clouds GMC: largest molecular clouds in ISM M ~ 105 M, but they gather together in complexes that have M ~ 106 M. Dense cores are where star formation happens CO emission (top) and intensity of ionized gas (bottom) around Orion (B. A. Wilson - T. M. Dame - M. R. W. Masheder - P. Thaddeus, 2005)
How It Began Solar nebula: gas between stars Solar nebula collapsed (see also Bennett et al. Chapter 8) Sun forms at center Some gravitational PE radiated away, rest heated cloud Angular momentum conservation disk forms Eventually T high enough for fusion H burning began Energy produced in core matched energy radiated from surface equilibrium Planets form in disk Orbital directions reflect spinning disk Why does the cloud begin to collapse? How long does the collapse take?
Star Forming in Orion More than 3000 stars are in this image amongst the gas and dust in the nebula (NASA, ESA, M. Robberto (Space Telescope Science Institute/ESA) and the Hubble Space Telescope Orion Treasury Project Team) (Credit: Mouser Williams)
Free-fall Timescale The collapse of a cloud of gas will proceed at the freefall timescale Consider a spherical cloud. The acceleration of a particle some distance x from the center is given by: Note: this assumes that the pressure is small enough that it does not affect the collapse. The solution to this will be of the form: Note: the free-fall time is independent of the size of the cloud. Initially, the collapse is slow, but accelerates toward the end.
Free-fall Timescale Assuming constant density during the collapse:
Jeans Critical Mass A gaseous sphere has two influences acting on it: gravity pulling it together, and kinetic energy of the constituent particles tending to cause it to fly apart. For a relatively static situation, there is a critical mass such that mass lower than this limit will disperse and mass above this limit will collapse. This is the Jeans critical mass. For a given density, one can also consider a Jeans radius.
Jean's Mass Consider a gas sphere Total thermal energy of the particles: Gravitational potential energy of the spherical configuration: Virial theorem applies if stable. Collapse means
Jean's Mass Rearranging: Introducing mass density: Jean's length: Jean's mass:
Jean's Mass Observations: Collapse for:, Jeans length decreases as density increases Typical molecular cloud (GMC core):
Alternate Derivation Compare sound speed to free-fall collapse Collapse occurs if: This gives the same form as the Jean's length
Collapse Large, cold cloud of gas (D ~ few ly) Collapse begins Gravity pulls cloud together Heating Gravitational potential energy released. ½ convert to heat; ½ radiated away (virial theorem) Sun forms at hot center Rotation Conservation of angular momentum Disk formation Easier to collapse along rotation axis than against it Collisions make orbits circular Outer part of disk thin cools rapidly. Temperature gradient forms.
Rotation and Collapse Angular momentum is conserved in rotating cloud Disk forms Smaller fragments of the cloud can collapse more easily Total angular momentum split between orbital angular momentum and rotational angular momentum of each cloud Fragmentation: can give rise to binary systems
Credit: Hubble Heritage Team (STScI/AURA), N. Walborn (STScI) & R. Barbß (La Plata Obs.), NASA http://antwrp.gsfc.nasa.gov/apod/ap030630.html Molecular cloud near Carina nebula. About 2 ly across, It is being eroded by radiation from nearby young stars.
Star Formation and the HR Diagram Jeans: free-fall initially Pressure becomes important and contraction proceeds on KH timescale 1/2 gravitational PE is radiated luminosity This phases is much longer than free-fall collapse H- opacity makes the star convective Hayashi track (Paxton et al. 2010)
Star Formation and the HR Diagram Hayashi track: boundary separating hydrostatic stars that can carry energy via convection (left) and those that have no means of effective energy transport (right) Move off Hayashi track when reactions kick in First part of pp-chain and conversion of 12C to 14N via CNO Low mass stars never have this happen Once hot enough, full H burning comences (Paxton et al. 2010)
Complications Rotation Conservation of angular momentum cloud spins up Collapse easier along rotation axis than across it protoplanetary disk forms Protostar can continue to grow via accretion Friction between orbits in disk causes inspiral Strong magnetic field can slow rotation
http://antwrp.gsfc.nasa.gov/apod/ap091222.html Credit: NASA, ESA, M. Robberto (STScI/ESA), the HST Orion Treasury Project Team, & L. Ricci (ESO)
Complications Jets Formation uncertain Aligned with spin axis Outflow liked channeled by magnetic field Binaries Cloud fragments during collapse Angular momentum carried in orbits
A disk around a protostar + a jet
Young Disk HL Tau: < 1 Myr; 140 pc ALMA (ESO/NAOJ/NRAO)
Slightly Older Disk HD 141569; 5 Myr, 98 pc PHY Stars and J. C. B. Papaloizou, A&A, 2004) (J. C.521: Augereau
An example of a disk: β Pictoris Beta Pictoris is young only about 10 million years old.
ESO/A.-M. Lagrange For the first time, astronomers have been able to directly follow the motion of an exoplanet as it moves to the other side of its host star. The planet has the smallest orbit so far of all directly imaged exoplanets, lying as close to its host star as Saturn is to the Sun. The above composite shows the reflected light on the dust disc in the outer part, as observed in 1996 with the ADONIS instrument on ESO's 3.6-metre telescope. In the central part, the observations of the planet obtained in 2003 and autumn 2009 with NACO are shown. The possible orbit of the planet is also indicated, albeit with the inclination angle exaggerated. http://www.eso.org/public/images/eso1024a/
Debris Disk Formalhautl 440 Myr, 7.6 pc At this stage, a planet is actively shaping the debris
Protostars Even though it is luminous, most of the radiation is in the IR Dust obscures our view
Initial Mass Function Low mass stars form more frequently than high mass stars Salpeter function: Salpeter, power is -2.35 Now we typically have different powers for low and high mass (Ivan Baldry)
T Tauri Stars Young (pre-main sequence) variable stars found near molecular clouds High activity, variability ~days Many have H, Ca II, K emission lines P Cygni profile has blue shifted absorption indicates disk Still contracting along vertical path in HR diagram < 2 solar masses Many have circumstellar disks Outburst from XZ Tauri, showing bubble of hot gas flowing out from the system John Krist (STScI), Karl Stapelfeldt (NASA Jet Propulsion Laboratory), Jeff Hester (Arizona State University), Chris Burrows (ESA/STScI) http://hubblesite.org/newscenter/archive/releases/2000/32/
T Tauri Stars (Carroll & Ostlie)
Next Step: Main Sequence Main-sequence begins when central temperature rises to ~10 million K hot enough for fusion Low mass stars remain on the main-sequence for 100 billion years High mass stars live for ~ 10 million years Planets can form in disk
Artists conception of the β-pic solar system http://www.nasa.gov/vision/universe/starsgalaxies/betapic.html (Credit: NASA/FUSE/Lynette Cook)
Outstanding Questions Do we need a trigger to initiate the collapse? Shock wave from nearby supernova? Stellar wind from young star? How much of the mass of the cloud winds up in stars? What is the initial mass function of stars? OB associations: unbound stars from a bound cloud how? Do high and low mass stars form in the same place? High mass stars may form is bursts
H II Regions Hot stars (O, early B) have significant flux in UV Able to ionize H H II region: region surrounding star where ionization takes place extends out to the point where ionization and recombination balance
H II Regions H II regions (red spots) in M51 (NASA, ESA, S. Beckwith (STScI), and The Hubble Heritage Team (STScI/AURA)) M33 (Chris Schur) http://antwrp.gsfc.nasa.gov/apod/ap061123.html
Exoplanets What about the disk?
Two Types of Planets (NASA)
Debris Main asteroids (green) Near-earth asteroids (red) Comets (blue squares) Trojan asteroids (blue dots) (Minor Planet Center, Gareth Williams)
Debris
Center of Mass Motion of the Sun (Wikipedia)
Planet Detection Overview Gravitational: observe motion of star around center of mass Directly: astrometric Doppler: radial velocity Photometric Transits Direct Imaging Microlensing All methods have their own strengths/weaknesses and biases
Planet Detection Methods (M. Perryman/Exoplanet.eu)
Astrometric Technique Look for wobble of parent star as it moves in its orbit around the galaxy Works very well for binary stars Wobble is very small for planets More pronounced for: Massive planet Larger orbits (but this requires longer observations) Closer stars
Radial Velocity Technique Up until recently (Kepler) most exoplanets were discovered this way Typical orbital velocities are 10s m/s Mass of the star is estimated via spectroscopy Observing a whole period tells use: Orbital period, and therefore semi-major axis Eccentricity of the orbit Minimum mass of the planet (M sin i) Number of planets Best for massive planets close to the star Higher v Shorter P
Radial Velocity Technique
Radial Velocity Technique 51 Pegasi b discovered this way Radial velocity curve for 51 Pegasi (from http://www.extrasolar.net/planettour.asp?planetid=1) Planet properties: M = 0.47 MJ P = 4.23 days distance = 0.052 AU
Radial Velocity Multiple planets Upsilon Andromeda system (Butler et al. 1999)
Transits No dynamical information Period: from time between transits Planet radius: from amount of dimming and knowledge of parent star Composition: spectra during and outside of transit Surface T: IR observations look for drop in flux
Transits Weaknesses: Strengths: Needs inclination of 90 degrees Biased toward planets with short periods Doesn't give mass Works for smaller (Earth-like planets) Get's radius Can be done with small telescopes Best of both worlds: transits + radial velocities Density can then be determined
Direct Detection High contrast IR helps (for young planets) Only a handful discovered this way
Direct Detection Planet around β Pictoris Note the debris disk http://antwrp.gsfc.nasa.gov/apod/ap081128.html Credit: ESO, A.-M. Lagrange (LAOG), et al.
Direct Detection HR 8799 4 known planets Large distances: challenge for formation theory (Marois et al. 2010, Nature) (Wikipedia)
Direct Detection Kuiper belt analogs (debris disks) Credit: NASA, ESA, P. Kalas, J. Graham, E. Chiang, E. Kite (Univ. California, Berkeley), M. Clampin ( NASA/Goddard), M. Fitzgerald (Lawrence Livermore NL), K. Stapelfeldt, J. Krist (NASA/JPL)
Imaging First image of a planet being formed around its star. Figure 1 Left: The transitional disk around the star LkCa 15. All of the light at this wavelength is emitted by cold dust in the disk. the hole in the center indicates an inner gap with radius of about 55 times the distance from the Earth to the Sun. Right: An expanded view of the central part of the cleared region, showing a composite of two reconstructed images (blue: 2.1 microns, from November 2010; red: 3.7 microns) for LkCa 15. The location of the central star is also marked. Credit: Kraus & Ireland 2011 Artist s conception of the view near the planet LkCa 15 b. Credit: Karen L. Teramura, UH IfA http://keckobservatory.org/news/first_close-up_view_of_a_planet_being_formed/
Kepler Mission Transit mission Launched 2009 Monitor 100,000 stars over 4 years Look for transits lasting hours, recurring on month-year timescales ~ 1 in 200 systems have proper orientation Planned to find ~ 50 earth's over the 4 year mission Start are more active than they thought originally Will find many more Jovian planets ~3000 candidates (orbits not confirmed) 134 confirmed exoplanets Extrapolation: 40 billion Earth-sized planets orbiting habitable zones around Sun-like a lower mass stars in our Galaxy Gyroscope failure resulted in original mission termination.
Kepler Mission http://kepler.nasa.gov/media/art.html#lomberg http://www.nasa.gov/mission_pages/kepler/multimedia/images/fullff IHot300.html
Kepler Candidates
Discovery History
Exoplanet Properties Very few known exoplanets have orbits beyond 5 AU Different databases give different #s of known (and wellknown) exoplanets
Exoplanet Properties Mass of Earth ~ 0.003 MJ Mass of Neptune ~ 0.05 MJ
Exoplanet Properties
Exoplanet Properties
Exoplanet Properties (from Udry and Santos, ARAA 2007)
Exoplanet Properties
Exoplanet Properties (from Udry and Santos, ARAA 2007)
Exoplanet Properties Metals are likely important in making planets
Competing Ideas on Planet Formation For Jovian planets, there is some uncertainty http://www.spacetelescope.org/images/html/opo0319f.html (NASA/ESA and A. Feild (STScI)
Core Accretion vs. Gravitational Instability (Baraffe et al. 2010) Core accretion difficulties Growth of cores takes longer than typical disk lifetimes (few Myr) Larger disks or lower opacity can help Turbulence in the disk may help Gravitational instability difficulties Requires disk to cool Only operates in massive disks and at large orbital distances (> 100 AU) May not be able to explain the higher metallicities in giant planets Gravitational instability works quickly young (1 Myr old) stars should already have gas planets
Core Accretion vs. Gravitational Instability HD 149026b is significantly denser than Saturn May have 70 M of heavy material Supports core-accretion model (Baraffe et al. 2010)
Brown Dwarfs (Baraffe et al. 2010) At the high end of the mass scale of exoplanets, with brown dwarfs Brown dwarfs are thought to form like stars, fragmentation of a giant molecular cloud Doesn't have to be in the plane of the planetary disk Planets form in the protoplanetary disk after star formation Either through core accretion or gravitational instability in the disk Planets should have higher amounts of heavy metals than brown dwarfs
Large Sizes (Baraffe et al. 2010) Some hot Jupiters have very large radii Close to parent star stellar irradiation Doesn't explain the entire effect Tidal heating? Atmospheric circulation (slows cooling, larger radius at given age)? Other physics (Ohmic heating?)
Comparing to Our Solar System Few terrestrial planets in these systems (selection effect) Much closer to star + higher eccentricities than our SS Hot Jupiters: composition probably same as our Jovian planets H/He Larger radii/lower density is due to closer orbit Moving Jupiter 0.05 AU would make it 50% bigger in radius Higher T (1000 K) rock-dust clouds! Cloud bands, strong winds (extreme heating + synchronous rotation)
Solar System Formation Problems How do you make a hot Jupiter? Nebular theory suggests that planet formation is a natural consequence of star formation Based on our solar system, Jupiters should be at large distances from their parent star Revisions based on exoplanet discoveries Jovian planets migrate inward
Planetary Migration Planet moving through disk creates density waves Type I migration Difference in torque from inner and outer waves transports angular momentum Operates for planets < 0.1 Jupiter masses Type II migration Planet clears a gap in the disk Gas flow in the disk moves the gap, planet moves with it
Planetary Migration
Encounters and Resonances Gravitational encounters eccentric orbits Two Jovian planets get close: 1 ejected, one spirals inward, elliptical orbit Small planetesimals ejected (to Oort cloud), Jovian planet loses orbital energy Happened in our SS Resonances Lead to eccentric orbits Can yield migration or ejection