Recent Work on the Lowest-Mass White Dwarfs

Colloquium, 45 min. November 2014: Keele University, Keele, UK.

Transcript

1. JJ Hermes
   
   University of Warwick

   View Slide

2. Mean Earth--Moon

   View Slide

3. Motivation and Outline
   •  Extremely Low-Mass (ELM, <0.3 M¤
   ) White Dwarfs (WDs)
   
   –  Set Galactic gravitational wave foreground
   
   –  eLISA verification binaries
   
   –  Progenitors of Galactic exotica: merged WDs and subdwarfs, AM CVn systems,

   View Slide

4. •  WDs: Burnt-out cores of all low-mass stars initially <8-10 M¤
   
   
   
   
   
   
   
   
   
   
   •  WDs are the endpoints of stellar evolution
   
   –  Their progenitors lost considerable mass
   
   White Dwarfs, the Quantum Dots
   

   View Slide

5. Kleinman  et  al.  2013,  ApJS,  204,  5
   •  All WDs discussed today have pure hydrogen atmospheres (DA)
   
   –  4/5 of WDs are DA; strong gravitational settling
   
   •  Estimate WD masses from observed Balmer line profiles: Teff
   /log(g)
   He-Core WDs
   CO-Core WDs
   ONe-Core WDs
   
   Mass Distribution of Known White Dwarfs
   

   View Slide

6. •  The Galaxy is not old enough for a
   single star to evolve into a

   View Slide

7. Latest  ELM  Survey  release:
   Brown  et  al.  2013,  ApJ,  769,  66
   •  ELM Survey: u-g, g-r color selection from Sloan Digital Sky Survey (SDSS)
   
   •  Discovery spectroscopy from

   View Slide

8. •  ELM Survey: SDSS u-g, g-r color selection
   
   •  Discovery spectroscopy from

   View Slide

9. Discovery spectroscopy
   determines the binary and
   atmospheric parameters
   
   blue-shifted
   red-shifted
   
   Brown  et  al.  2012,  ApJ,  744,  142
   T
   eff
    =  10,540  ±  170  K
   log(g)  =  6.01  ±  0.06
   P
   orb
    =  87.996  ±  0.006  min
   K
   1
    =  508  ±  4  km  s-­‐‑1
   M
   2
    >  1.10  M¤    
    if  M
   1
    =  0.17  M¤
   t
   merge
    <  170  Myr
   J1741+6526: An 88-min WD+WD binary
   

   View Slide

10. Asin(φ)
    = 0.50 ± 0.08 %
    
    
    Doppler beaming
    
    Acos(2φ)
    = 1.30 ± 0.08 %
    
    
    Ellipsoidal variations
    
    J1741+6526: An 88-min WD+WD binary
    •  Follow-up photometry yields
    further physical constraints
    
    

    View Slide

11. •  Doppler beaming: Radiation is beamed toward our line of sight,
    proportional to how fast the source is moving (V):
    
    
    
    
    
    •  (Also a small factor for the

    View Slide

12. •  Ellipsoidal variations: Changing projected area
    of a tidally distorted star
    
    •  Tidal bulge rotates once per orbit
    
    –  We see its oblique (fat) side

    View Slide

13. J0751-0141: A New Eclipsing WD+WD Binary
    Kilic  et  al.  2014,  
    MNRAS,  438,  L26
    P
    orb
     =  115.22  min
    
    
    3.2% rel. amplitude

    View Slide

14. Yellow: Our
    eight new
    tidally
    distorted
    ELM WDs
    
    
    
    Black:
    Eclipsing
    WD+WD
    binaries
    
    
    
    Pink:
    Eclipsing

    View Slide

15. Some Open Questions Regarding ELM WDs
    – CNO flashing episodes and HR-diagram loops
    
    – Hydrogen-layer masses in He-core WDs
    
    – The ubiquity of metals in the lowest-gravity WDs
    
    – Tidal torques on binary inspiral
    
    

    View Slide

16. •  Lowest-mass WDs
    (≤0.18 M¤
    ) have

    View Slide

17. ELM WDs and Predicted CNO Flashes
    Althaus  et  al.  2013,  A&A,  557,  A19
    •  Two low-mass WDs of different masses often cross the same points in a
    T
    eff
    —log(g)  diagram
    
    •  There is a non-uniqueness to using T
    eff
    ,log(g) for ELM WD mass
    
    

    View Slide

18. ELM WDs and Predicted CNO Flashes
    Althaus  et  al.  2013,  A&A,  557,  A19
    •  For example, take a 10,000 K, log(g) = 6.60 WD:
    
    

    View Slide

19. Some Open Questions Regarding ELM WDs
    – CNO flashing episodes and HR-diagram loops
    
    – Hydrogen-layer masses in He-core WDs
    
    – The ubiquity of metals in the lowest-gravity WDs
    
    – Tidal torques on binary inspiral
    
    

    View Slide

20. Discovery of Pulsations in Low-Mass WDs
    Hermes  et  al.  2012,  ApJ,  750,  L28
    Hermes  et  al.  2013,  ApJ,  765,  102
    Hermes  et  al.  2013,  MNRAS,  436,  3573
    Kilic  et  al.  2015,  MNRAS,  446,  26
    •  CNO flashes erode the hydrogen
    layer mass of ELM WDs
    
    •  An observational test would come
    from pulsating WDs: asteroseismology
    
    •  Since October 2011 we discovered
    the first five pulsating low-mass,

    View Slide

21. Pulsating Hydrogen-Atmosphere (DA) WDs
    •  Global g-mode pulsations driven by a hydrogen partial ionization zone
    
    

    View Slide

22. G117-B15A: A 0.59 M¤
    Pulsating CO-Core WD
    •  Stable pulsating WD
    
    •  dP/dt ~ 4 x 10-15 s s-1
    
    •  Main pulsation modes:
    
    –  P1
    = 215.2 s
    
    –  P2
    = 304.1 s
    
    –  P3
    = 270.5 s
    
    •  Can multiply the star’s
    frequencies by 300,000 to
    convert to audible range:
    
    Kepler  et  al.  2005,  ApJ,  634,  1311
    target
    
    comparison
    
    

    View Slide

23. J1614+1912: A 0.20 M¤
    Pulsating ELM WD
    •  The pulsating ELM WD
    with the shortest-period
    variability
    
    •  Main pulsation modes:
    
    –  P1
    = 1262.7 s
    
    –  P2
    = 1184.1 s
    
    •  Scaling frequencies by
    300,000 to an audible
    range:
    
    Hermes  et  al.  2013,  MNRAS,  436,  3573
    target
    
    comparison
    
    

    View Slide

24. J2228+3623: A 0.16 M¤
    Pulsating ELM WD
    •  The pulsating ELM WD
    with the longest-period
    variability
    
    •  Main pulsation modes:
    
    –  P1
    = 4181 s
    
    –  P2
    = 3252 s
    
    –  P3
    = 6229 s
    
    •  Scaling frequencies by
    300,000 to an audible
    range:
    
    target
    
    comparison
    
    Hermes  et  al.  2013,  MNRAS,  436,  3573
    

    View Slide

25. J1112+1117: A 0.17 M¤
    Pulsating ELM WD
    Hermes  et  al.  2013,  ApJ,  765,  102
    •  Main pulsation modes:
    
    –  P1
    = 2258.5 s
    
    –  P2
    = 2539.7 s
    
    –  P3
    = 1884.6 s
    
    –  P4
    = 2855.7 s
    
    –  P5
    = 1792.9 s
    
    •  Scaling frequencies by
    300,000 to an audible
    range:
    
    target
    
    comparison
    
    

    View Slide

26. Full Seismology Will Reveal ELM WD Structure
    •  We are currently only able to
    qualitatively match the periods
    to WD models
    
    •  Near to having a large grid of
    He-core WD models with
    different hydrogen layer masses
    to perform asteroseismology
    
    Van  Grootel  et  al.  2013,  ApJ,  762,  57
    MH
    /M* = 10-4
    
    MH
    /M* = 10-2
    
    J1840
    
    J1518
    
    J1518
    
    J1112
    
    Theoretical periods for ell=1 g-modes modes vs.
    Observed Periods
    
    

    View Slide

27. Some Open Questions Regarding ELM WDs
    – CNO flashing episodes and HR-diagram loops
    
    – Hydrogen-layer masses in He-core WDs
    
    – The ubiquity of metals in the lowest-gravity WDs
    
    – Tidal torques on binary inspiral
    
    

    View Slide

28. Lowest-Gravity WDs All Show Metals
    Hermes  et  al.  2014,  MNRAS,  444,  1674
    

    View Slide

29. •  Roughly 1 in every 2-3 WDs we find has some metal pollution
    
    •  Metals should settle out of the high-surface-gravity atmosphere very
    quickly (of order days)
    
    •  Consensus: Metals are from accreted, tidally disrupted debris
    
    •  Abundances match bulk

    View Slide

30. •  Ca II lines phase with the ~288 km/s RV of the Balmer lines
    
    –  Metals are photospheric, not interstellar
    
    •  We obtained an HST/COS

    View Slide

31. •  This ELM WD is carbon deficient, just like planetary debris
    
    •  BUT: Oxygen abundance inconsistent with rocky accretion:

    View Slide

32. GALEX J1717: A 5.9-hr, Metal-Rich He-WD+WD
    R
    1
     =  0.093  ±  0.013  R¤
    
    i  =  86.9  ±  0.4  deg
    P
    orb
     =  5.90724895(41)  hr
    -­‐‑20
    v
    rot
     =  50+30  km  s-­‐‑1
    
    P
    rot
     =  2.3+2.0  hr
    Hermes  et  al.  2014,  MNRAS,  444,  1674
    •  Prot
    < Porb
    but not yet formally significant
    
    •  Direct test of tidal synchronization!
    
    -­‐‑1.0
    

    View Slide

33. Some Open Questions Regarding ELM WDs
    – CNO flashing episodes and HR-diagram loops
    
    – Hydrogen-layer masses in He-core WDs
    
    – The ubiquity of metals in the lowest-gravity WDs
    
    – Tidal torques on binary inspiral
    
    

    View Slide

34. phase = 0
    
    •  We detected eclipses in April 2011
    
    •  This is the most compact detached binary
    system currently known!
    
    J0651+2844: A 12.75-min WD+WD Binary
    Brown  et  al.  2011,  ApJ,  737,  L23
    

    View Slide

35. J0651+2844: A 12.75-min WD+WD Binary
    Average distance between the Earth and the Moon: 384,400 km
    
    

    View Slide

36. (from Phase 0 to Phase 1 is 12.75 minutes)
    
    Hermes  et  al.  2012,  ApJ,  757,  L21
    P
    orb
     =  765.20644(95)  s
    K
    1
     =  616.9  ±  5.0  km  s-­‐‑1
    i  =  86.3  ±  1.0  deg
    T
    eff,1
     =  16,340  ±  260  K
    M
    1
     =  0.252  ±  0.04  M¤
    T
    eff,2
     =  10,370  ±  360  K
    M
    2
     =  0.50  ±  0.04  M¤
    J0651+2844: A 12.75-min WD+WD Binary
    

    View Slide

37. Orbital Decay in J0651+2844
    After just 13 months we confirmed orbital decay from gravitational radiation.
    
    Hermes  et  al.  2012,  ApJ,  757,  L21
    

    View Slide

38. We expect dPorb
    /dt = (-0.26 ± 0.05) ms/yr and observe (-0.2834 ± 0.0039) ms/yr!
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    – a 1.4% measurement!
    
    
    Orbital Decay in J0651+2844
    

    View Slide

39. •  This 12.75-min WD+WD binary is decaying > 3.5 times faster than the
    7.75-hr Hulse-Taylor binary pulsar, which was the first indirect detection of
    gravitational radiation (1993 Nobel prize in physics)
    
    Weisberg  et  al.  2010,  ApJ,  722,  1030
    J0651+2844
    PSR B1913+16
    
    dP/dt = -0.283 ms/yr
    dP/dt = -0.076 ms/yr
    
    Orbital Decay in J0651+2844
    

    View Slide

40. •  Gravity bends space; it effectively
    determines geometry of space
    
    •  General Relativity: Any mass in
    nonuniform, nonspherical motion
    emits gravitational radiation
    
    •  Ripples in space-time caused by
    gravitational radiation carry away
    energy
    
    •  This is an energy leak and acts

    View Slide

41. •  J0651+2844 is an excellent verification sourcefor direct detection of

    View Slide

42. •  Tidal torques should increase the rate of
    orbital decay in J0651+2844
    
    –  Additional angular momentum is lost from the
    orbit to spin-up the WDs to remain
    synchronized, leading to >5% faster rate of
    orbital decay (e.g.,  Piro  2011,  ApJ,  740,  L53;  Fuller  &  
    Lai  2012,  MNRAS,  421,  426)
    The Fate of the WDs in J0651+2844
    

    View Slide

43. Conclusions
    •  Extremely Low-Mass (ELM, <0.3 M¤
    ) White Dwarfs (WDs) constrain
    the endpoints of stellar and binary evolution
    
    •  “Low-Mass White Dwarfs Need Friends”
    
    –  Close companions provide many ways to observationally constrain systems
    
    •  Pulsations allow us a new way to explore He-Core, ELM WD Interiors
    
    •  ELM WDs provide a unique test for tidal effects on binary inspiral
    
    •  The first directly detected gravitational waves and confirmed EM
    counterpart systems will likely be ELM WDs
    
    D. Berry, GSFC!
    

    View Slide

44. View Slide

45. Hermes  et  al.  2014,  ApJ,  792,  39
    

    View Slide

46. Corsico  &  Althaus.  2014,  A&A,  569,  106
    

    View Slide

47. View Slide