Lecture 3: Formation of Low Mass Stars and Planets

1. Lecture 3: Formation of Low Mass Stars and Planets Neal Evans The University of Texas at Austin and KASI

2. Part 4: Core to Star + Disk • The idea of a Core • Some simple theory • Evolutionary Stages and SED Classes • Timescales • Tests of theory

3. Dense Cores • Observations find regions of strongly enhanced density, decreased turbulence • Masses comparable to stellar masses • Identified as birthplaces of individual stars (single, binary, or multiple stars) • Gravitational collapse

5. Central Condensation Radial profile (mm continuum emission vs impact parameter) Observed (points) vs model of a Bonnor-Ebert sphere (line) Centrally condensed with density approaching n ~ r–2 Models of Bonnor-Ebert spheres with increasing central condensation

6. Part 4: Core to Star + Disk • The idea of a Core • Some simple theory • Evolutionary Stages and SED Classes • Timescales • Tests of theory

7. Inside-out Collapse • F. Shu derived basic model for collapse • Initially stabilized (magnetic, turbulence?) • Gradually relaxes to n ~ r–2 density profile • Supported by thermal pressure • Collapse begins on inside and propagates out

8. The Equations • rinf = cst infall inside this radius • dM/dt = cs3/G Mass infall rate • n ~ r–2 ; vinf = 0 for r > rinf • r(r) = cs2/[2p G] r–2 • n ~ r–1.5 ; vinf ~ r–0.5 for r < rinf • r(r) = dM/dt/[4p(2GM)0.5] r–1.5 • Shu (1977), Shu, Adams, Lizano (1987)

9. Evolution of velocity/density

10. Infall to Star: Luminosity • Calculate energy released by accretion • L = GM*/R* x dMacc /dt • Luminosity of star dominated by accretion while material falling in • Heats up dust in infalling envelope • Td ~ r–0.4 for likely dust properties • Dust heats up gas in envelope • Collisions of gas with warmer dust

11. Density and Temperature(s)

12. Angular Momentum • Measure of tendency to rotate • J = mvr • Angular momentum is conserved • J = constant • As gas contracts (r smaller), v increases • Faster rotation resists collapse • Gas settles into rotating disk • Protostar adds mass through the disk

13. The Equations • RC = G3M3W2/16 cs8 • M is mass of central star + disk • W is angular speed of initial core • Inside RC, rotation > infall speed • Disk forms, grows ~ t3 • Terebey, Shu, and Cassen (1984)

14. Quadrant of solution Gray scale is density; arrows indicate flow velocity Material lands off center, forms disk

15. Angular Momentum Problem • Even if a tiny amount of rotation initially • Conservation of angular momentum would make star rotate faster than speed of light • Magnetic fields used to slow and remove angular momentum • Rotation + B launches a fast wind/jet • Jet clears a cavity and sweeps up infalling material into bipolar outflow

16. Cartoon of the Protostar Features: Infalling envelope Rotation Disk Jet Bipolar outflow R. Hurt, SSC

17. The Bipolar Jet

18. An Example Model Density Temperature BHR71 model (Yang et al. 2017)

19. Radial Profile Radial profile now depends on angle. q = 90 is equatorial, so does not intersect outflow cavity. Others intersect at radii dependent on the angle. BHR71 model (Yang et al. 2017)

20. Part 4: Core to Star + Disk • The idea of a Core • Some simple theory • Evolutionary Stages and SED Classes • Timescales • Tests of theory

21. Formation of a Star Single isolated low-mass star outflow n~105-108 cm-3 T~10-300 K Stages infall Factor 1000 smaller Protostar with disk Cloud collapse t= 5 x105yr t=0 C Collapse of a cold, dense core Matter falling onto forming star releases energy To avoid spinning too fast, matter forms a disk, which feeds matter to the star. As the star grows, temperature rises in infalling material Formation planets t=106-107 yr Solar system t>108 yr

22. From a Disk to Planets outflow Formation planets Solar system Protostar with disk Cloud collapse t=0 t=105 yr Dusty disk survives about 2-3 Myr; planets form from dust Formation planets t=106-107 yr Solar system t>108 yr Spitzer probes dust at temperatures between 100 and 1500 K.

23. Standard evolutionary scenario single isolated low-mass star All SEDs from Dunham et al. (2013), PPVI review chapter ? Classes Candidate FHSC Class 0 protostar Prestellar core Wavelength (μm) Figure adapted from McCaughrean, unpublished, by A. Stutz outflow n~104-105 cm-3 T~10 K n~105-108 cm-3 T~10-300 K ? Stages infall First hydrostatic core t = 0 t = 105 yr (?) Protostar with disk Core collapse

24. Solar system Standard evolutionary scenario single isolated low-mass star SEDs from Dunham et al. (2013), PPVI reviewchapter SED from Fischer et al. (in prep.) ? Classes Class I protostar Flat spectrum Class II YSO Wavelength (μm) Figure adapted from McCaughrean, unpublished, by A. Stutz outflow n~105-108 cm-3 T~10-300 K Envelope dissipation? Stages infall t = 105 yr (?) t = 106-107 yr Protostar with disk Formation of planets

25. Class 0 Class I Class II How do we classify protostars? Based on the shape of the observed SED α = dlog(λSλ) dlogλ • SED slope (α method): original criteria for Classes • (Lada 1987; Greene et al., 1994) • LSMM/LBOL: added later to identify Class 0 (Andre et al., 1993, also Maury et al., 2011) • Bolometric temperature (Myers & Ladd, 1993): the temperature of a black body with the same flux weighted mean frequency as the observed SED (see also Greene et al.,1994). Class 0 LSMM/LBOL > 0.5% TBOL ≤ 70 K Class I α ≥ 0.3 70 K < TBOL ≤ 670 K Flat -0.3 ≤ α < 0.3 Class II -1.6 ≤ α < -0.3 670 K < TBOL ≤ 2800K Class III α < -1.6 TBOL > 2800 K All SEDs from Dunham et al. (2013), PPVI review chapter

26. ENVELOPE EVOLUTION? How do we think they evolve? Class 0 LSMM/LBOL > 0.5% TBOL ≤ 70 K Class I α ≥ 0.3 70 K < TBOL ≤ 670 K Flat -0.3 ≤ α < 0.3 Class II -1.6 ≤ α < -0.3 670 K < TBOL ≤ 2800K Class III α < -1.6 TBOL > 2800 K Class 0 Class I Class II All SEDs from Dunham et al. (2013), PPVI review chapter

27. Part 4: Core to Star + Disk • The idea of a Core • Some simple theory • Evolutionary Stages and SED Classes • Timescales • Tests of theory

28. I: Flat: II: III: IF time is the only variable AND IF star formation continuous for t >> t(II) THEN t(Class) = t(II)*N(class)/N(II) Caveats: Class III census incomplete Class III not included in timescale Depends on how  is calculated Class 0 mixed with Class I t(II) may be longer; this was based on half life of IR excess in clusters, but stellar ages may be longer (PPVI) If t is not >> t(II), in build up phase and durations are shorter. Timescales for Classes

29. Numbers of YSOs and lifetimes Average half-life of Class 0+I: 0.42 to 0.54 Myr assuming a 2 Myr Class II half-life Table from Dunham et al., 2013, PPVI review chapter

30. Part 4: Core to Star + Disk • The idea of a Core • Some simple theory • Evolutionary Stages and SED Classes • Timescales • Tests of theory

31. Comparison to Shu model • Assume inside-out collapse at 0.19 km/s • Sound speed at 10 K • In 0.54/2 Myr, rinf = 0.054 pc • Consistent with some sizes • Mean separation in clusters 0.072 pc (Gutermuth) • At dM/dt = 1.6 x 10–6Msun/yr, M* ~ f 0.86 Msun • If f ~ 0.3, get 0.26 Msun ~ modal mass • Infall rate is right to build star in allowed time • Consistent with assumptions, most data • Picture holds together, except…

32. The Luminosity Problem! M. M. Dunham et al. 2010

33. Many are under-luminous The ACCRETION rate has to be slower Predicted L = GM(dM/dt)/R= 1.6 Lsun for standard (Shu) accretion onto M = 0.08 Msun, R = 3 Rsun. Most (59%) are below this. M. M. Dunham et al. 2010

34. Episodic Accretion • Infall rate like Shu, but accretion rate highly variable • Kenyon and Hartmann (1995) suggested this to solve luminosity problem (IRAS) • Exacerbated by Spitzer data • Simulations show it (Vorobyov and Basu 2005, 2006) • Infall from envelope to disk is not obviously synchronized with accretion from disk to star

35. Direct Evidence for Episodic Accretion • Luminosity Variations (e.g., FU Orionis) • VeLLOs (L<0.1 Lsun), much less than prediction for standard accretion onto BD/star • Outflow morphologies suggesting multiple ejection events (e.g., HH 211) • Comparison of L(now) with <L(t)> • Outflows trace history of ejection, hence accretion • Careful analysis of several sources gives strong evidence for L(now) < <L(t)> • Dunham et al. 2006, 2010

36. Improved fit to BLT Data Shading indicates time spent in that cell of BLT diagram in (more sophisticated) episodic models Dunham & Vorobyov 2012

37. And 1D Distributions Dunham & Vorobyov 2012

38. Consequences of Episodicity • The connection between Classes and Stages becomes tenuous • The luminosity is not an indicator of stellar mass until nuclear burning dominates • (Lacc ~ M*dMacc/dt) • The initial conditions for planet formation may be determined by time since last episode of disk instability

39. From Dust to Planets • Terrestrial (Earth-like) planets • Built from microscopic dust grains in the disk around the forming star • About 5 x 1038 dust particles needed to make Earth • The early steps (up to km size) have to be complete in about 3 Myr

40. First Step: Accretion of Dust Grains Fig. From talk by Jurgen Blum

41. Second Step: Dust settles to midplane Lowest GPE Midplane Williams and Cieza, Annual Reviews

42. Step 3: Dust in midplane grows to rocks, boulders, … Artist’s conception: Vega

43. Step 4: From Boulders to Planets • Boulders grow to planetesimals • Planetesimals collide, grow larger • Some dust returned in collisions • Icy dust in outer part of disk • Builds bigger, icier planets (Uranus, Neptune) • Internal heat turns ice to gas • If rock-ice core massive enough • Gravitational collapse of gas • Gas giants with ring/moon systems (Jupiter, Saturn)

44. Planetesimals Collide ESA

45. Dust: Evolution of Mineralogy Forsterite (Mg2SiO4) 8 Young stars show this feature, indicative of heating HD100546 Sturm et al. 2010

46. Ice: Crystalline H2O ice • Ice feature around 60-70 microns is broad • Very careful calibration of the spectral response function is needed to see it • We now believe it in a few disks • See also McClure et al. (2012) in GQ Lup

47. Water Ice in the Outer Disk __________ ^Bouwman, Henning et al. In prep. Min et al. 2016 Water Ice Ice is crystalline, best explained by collisions of icy planetesimals. ()

48. Water Vapor in Inner Disk

49. Rock-Ice Objects in Kuiper Belt Pluto is best studied example. Rock and water ice, with nitrogen ice surface. Rocky stuff ~70% Icy stuff ~ 30%

50. The Earth is a dry planet The Earth is a wet planet USGS = 6 million comet Halleys 1 Earth ocean = 0.00023 Mearth