Direct detection of substellar and planetary companions to white dwarfs

1. Direct detection of substellar and planetary companions to white dwarfs Matt Burleigh

2. with thanks to: • Jay Farihi, Sarah Casewell, Martin Barstow, Richard Jameson, Katherine Lawrie (Leicester) • Paul Steele (MPA Garching) • Francesca Faedi (Belfast) • Paul Dobbie (AAO) • Fraser Clarke (Oxford), Emma Hogan (Gemini South) • Ted von Hippel, Fergal Mullally (Kepler) • Hans Zinnecker, Suzanne Friedrich

3. Overview • IR searches for substellar companions • UKIDSS survey results • Direct imaging: into the planetary regime • Ground-based: DODO survey • Space-based: Spitzer survey • Latest results: a close WD+ Tdwarf binary in an open cluster • A transit survey of WDs with SuperWASP

4. Direct imaging of planets around white dwarfsBurleigh, Clarke & Hodgkin 2002, MNRAS, 331, L41 • Ideal targets for direct imaging (and spectroscopy) of exoplanets: • Improve contrast problem: WDs up to ~104x fainter than their progenitors • Improve resolution problem through widening of orbits • M(MS) / M(WD) (Jeans 1924) • Indirectly study planetary systems around WD progenitors: early-type stars (B,A,F), intermediate masses (1.5<M<8Msun) • White dwarfs common in solar neighbourhood • White dwarf ages are well constrained • Planetary companions of WDs will be old, mature gas giants • 200 < Teff < 600K • > few x 108 years old • Detectable at ages up to several Gyr • Most direct imaging programmes target young stars

5. The end of our solar system -4000 -2000 0 Solar Radii

6. Evidence for old planetary systems • Planets have been found by radial velocity technique around evolved giant stars • 9% of stars 1.3<M<1.9Msun have planets (Johnson et al. 2007) • >3% of stars M>1.8Msun have planets >5MJup (Lovis and Mayor 2007) • White dwarfs have been identified as wide companions of planet-hosting stars • eg CD-38 10980 (Mayor et al. 2004), eps Ret (Raghavan et al. 2006, Chauvin et al. 2006, Farihi et al. 2011) • Growing number of brown dwarf companions in close and wide orbits • eg Maxted et al. 2006, Steele et al. 2011, Girven et al. 2011, Debes et al. 2011 • Metal-rich circumstellar dust and gas disks discovered around white dwarfs • See various papers by eg Jay Farihi & collaborators, Boris Gaensicke & friends

7. Dust disks around white dwarfs • A dozen WDs are known to be surrounded by dust disks • Disks identified as near-IR and mid-IR excesses, 500oK < T < 1200oK • Disks within a few solar radii of the WDs • Material within the disks is being accreted onto the WD atmosphere • Finally explains WDs with metal-polluted atmospheres

8. Dust disks around white dwarfs • Disks dominated by silicates • Disk forms from tidal disruption asteroid • Mass of GD362 disk suggests Mars-sized body • Suggests up to 20% of WDs may have rocky companions • If all polluted WDs have accreted terrestrial material at some point • But asteroids have to be moved by something: • Where are the perturbing giant planets?

9. HR8799 as a white dwarf… • HR8799: A5V 1.2-1.8Mjup 39pc 30-160Myr old • Planets: 7Mjup @ 68AU, 10Mjup @ 38AU, 10Mjup @ 24AU • Will evolve after 1.75Gyr to a 0.58Msun white dwarf • Planet orbits expand by factor ~3 to ~200AU,~120AU & 75AU • Let WD cool to 10,000K over ~0.5Gyr….system age now 2.25Gyr… • 10Mjup planet will be J=23.8 @39pc, or J=22.4 @ 20pc • How common are HR8799-like systems?

10. IR searches for substellar companions:UKIDSS survey results • Historical searches for brown dwarf companions to WDs going back to 1980s • egProbst(1983), Zuckerman & Becklin (1987), Wachter et al. (2003), Farihi et al. (2005), Dobbie et al. (2005), Tremblay & Bergeron (2007) • GD165B (L4), GD1400B (L6/7), WD0137-349B (L8) • UKIDSS survey • ~dozen new BD candidates • Several confirmed (eg PHL5038, Steele et al. 2009, A&A, 500, 1207) • WD+Lfraction >0.4+/-0.3% • WD+T fraction >0.2% • WD+BD fraction >0.5+/-0.3% • Steele et al. 2011, MNRAS, 416, 2768 • See also similar study by Girven et al. 2011, MNRAS, 417, 1210 • Parallel study for very wide binaries in UKIDSS discovered a WD+T4.5 (Day-Jones et al. 2009)

11. Direct Imaging Searches Around White Dwarfs: The DODO project Degenerate Objects around Degenerate Objects With Emma Hogan (now Gemini South) and Fraser Clarke (Oxford) The idea: Burleigh et al. 2002, MNRAS, 331, L41 Results: Burleigh et al. 2008, MNRAS, 386, L5 and Hogan et al. 2009, MNRAS, 396, 2074

12. ^ ~90” V • ~40 white dwarfs within 20pc, ages <2-3Gyr • J band survey, depths J=23-24 < ~120” >

13. Van Maanen’s Star (d=4.4pc) • DODO ground-based J-band imaging: • No companion found >7+/-1MJup (300+/-20K) at separations from 3”-45” / 15AU – 200AU • Separations around progenitor: 3 - 40AU • Spitzer mid-IR photometry: • No unresolved companions > 10+/-2 Mjup • Burleigh et al. 2008, MNRAS, 386, L5 • No resolved companions > 4MJup • Farihi et al. (2008)

14. A candidate? • Candidate mass ~7MJup WD

15. No candidate :( • 3rd epoch in 2008 refuted candidate

16. Limits on giant planets at white dwarfs: mass • <5% of WDs have resolved substellar companions above deuterium burning limit (13MJup) • <7% of WDs have resolved substellar companions >10MJup • <1/3rd of WDs have resolved substellar companions >6MJup

17. Limits on giant planets at white dwarfs: temperature • <4% of WDs have resolved substellar companions >500K • <7% of WDs have resolved substellar companions >400K

18. Maximum and minimum orbital separations probed by DODO

19. Comparison with Spitzer limits on unresolved companions(Farihi et al 2008, 34 targets)

20. Spitzer warm mission programme • Repeat observations of ~90 white dwarfs originally observed 2004/5 • Prog ID: 60161 • Title: “Cool, spatially resolved substellar & exoplantary analogues at white dwarfs” • PI: Burleigh, co-Is Farihi, Steele, Mullally, von Hippel • Look for common proper motion companions • 4.5micron band only

21. Spitzer 4.5micron imageGJ3483 (LTT3059 / WD0806-661) I am the white dwarf 130” / 2500AU I maybe a planet… or a brown dwarf

24. Proper motion • PM error +/-25mas/yr WD Companion

25. Candidate parameters • WD • 0.62Msun, • Progenitor mass 1.8-2.4Msun • Total age 1.2-2.0Gyr • Distance 19.2pc • Candidate • 4.5micron mag = 16.75+/-0.08 • 6-10MJup • 310-380K • Binary • Projected separation 130” / 2500AU • Original separation 700AU

27. Candidate parameters • WD • 0.62Msun, • Progenitor mass 1.8-2.4Msun • Total age 1.2-2.0Gyr • Distance 19.2pc • Candidate • 4.5micron mag = 16.75+/-0.08 • 6-10MJup • 310-380K • Binary • Projected separation 130” / 2500AU • Original separation 700AU

28. Y band observation • AAT / IRIS2 2hours exposure on 25th March 2011 • No detection • Sensitivity • Recover 50% of injected fake stars at 21.1

29. J-band observations • Rodriguez et al. 2011,ApJ, 732, 1222; Luhman et al 2012, ApJ, 744, 135 • No detection to J=23.9 (3 sigma) • J-[4.5] > 7 • Redder than any known T dwarf – a Y dwarf? • Suggest mass 6-9MJup and 310K < Teff < 350K

30. Planet or brown dwarf? • Is GJ3483B a brown dwarf or a planet? • Forget deuterium burning limit as the discriminator • can we classify by formation mechanism? • Original projected separation ~700AU • Too large for core accretion in a disk • Suggests disk fragmentation -> BD • Rodriguez et al. 2011 • But unstable, eccentric orbits expected in end states of stellar evolution • Debes & Sigurdsson 2002, Villaver & Livio 2007, Veras et al. 2011 • Disk of 2Msun star may be massive enough to make 6MJup companion • Progenitor could be an HR8799-like system • A5V+7MJup @68AU + 10MJup@38AU + 10MJup@24AU

31. Spitzer survey completeness

32. Spitzer survey completeness

33. Spitzer survey completeness Black: Spitzer resolved Red: DODO resolved (Hogan et al. 2009) Green: Spitzer unresolved (Farihi et al. 2008)

34. A WD+T dwarf close binary in Praesepe • RV & V band variability (4.2hr) in a WD in Praesepe cluster • M sin i > 25MJup • Non-detection in NIR and mid-IR (Spitzer IRAC) limits spectral type <T5 • At age of Praesepe (625Myr) M<30MJup • Strict age limit gives tight constraints on common envelope evolution parameters • Original orbit ~2AU (max extent of AGB envelope) • WD progenitor 3.5Msun/B9V • How did original binary form? • Most likely, capture early in cluster lifetime • Is this a significant method to populate the “brown dwarf desert”? • For more details see poster by Sarah Casewell

35. A search for eclipsing and transiting planets with SuperWASP • SuperWASP world’s most successful ground-based transit survey (>80 planets so far) • Has observed ~200 WDs since 2004 • 1% photometry to V=13, detection limit V~15 • No eclipsing or transiting companions detected • Placed (weak!) limits on incidence of such objects in close orbits around these white dwarfs • accounting for observing efficiency & co-variant noise in data • Limiting factors: • cadence: 8min for WASP v transit times of 1-few mins • unknown frequency of close planetary companions • Survivors of common envelope evolution? • 2nd generation planets? • Faedi, West, Burleigh, Goad, & Hebb, (2011), MNRAS, 410, 899 and various conference proceedings since 2007 • Please read the poster for more details

36. Open questions, future directions • How common are GJ3483-like objects? • More direct imaging searches for wide companions • What are their formation mechanisms? • Disk fragmentation? • Core accretion and subsequent ejection? • Where are the perturbers that help create dust disks around white dwarfs? • Mid-IR photometric searches; adaptive optics & HST • What is the lowest mass that can survive CE evolution intact? • Transit/eclipse searches • What is the orbital period distribution for substellar companions? • Are there “deserts”? • Can rocky planets exist in close orbits to WDs? • Transit searches • Can 2nd generation planets form? • Hot, young gas giants, metal-rich terrestrial planets?

37. The lowest mass companions to WDs