Double White Dwarf Mergers and the Formation of R CrB Stars

Size: px

Start display at page:

Download "Double White Dwarf Mergers and the Formation of R CrB Stars"

Randall Bell
5 years ago
Views:

1 Double White Dwarf Mergers and the Formation of R CrB Stars Josiah Schwab Hubble Fellow, UC Santa Cruz 13 September 2018

On the way to their final fates, double WD systems evolve through multiple phases. Lagrangian Hydrodynamics Eulerian Hydrodynamics Stellar Evolution (MESA) z [10 9 cm] t = 00000.

2 On the way to their final fates, double WD systems evolve through multiple phases. Lagrangian Hydrodynamics Eulerian Hydrodynamics Stellar Evolution (MESA) z [10 9 cm] t = s R [10 9 cm] Temperature [10 8 K] z [10 9 cm] t = s R [10 9 cm] Temperature [10 8 K] minutes hours yr

3 I ve applied this approach broadly over the possible range of WD+WD mergers. ONe WD M2 CO WD Collapse to NS after giant phase Schwab et al. (2016) Collapse to NS (but when?) Schwab et al. (in progress) He WD Single hot subdwarf stars Schwab (2018) R Coronae Borealis stars Schwab et al. (in prep) Collapse to NS during giant phase Brooks et al. (2017) He WD CO WD ONe WD M 1

4 Overview Models of RCrB Stars Models of their more massive cousins Summary

5 White dwarf mergers seem to provide a natural explanation for the formation of these objects. He He + CO = CO H-burning products He-burning products see e.g., Jeffery, Karakas, & Saio (2011) In addition, there can be further H and He burning during the merger.

6 The high nitrogen abundance requires additional CNO processing during the merger. from Asplund et al. (2000)

7 Getting the detailed CNO abundances right is challenging. You can reach the right conditions to make 18 O. Clayton et al. (2007), Staff et al. (2012) 14 N gets destroyed in making the 18 O, so you need to make even more 14 N. Menon et al. (2013), Zhang & Jeffery (2014) The outer layers of the CO WDs are generally oxygen-rich. During the merger, it is difficult not to bring up a lot of 16 O. Does this require the presence of a large He-buffer layer? Staff et al. (2012, 2018)

8 Mixing is essential but also uncertain. Zhang & Jeffery (2014) include convective mixing. Menon et al. (2013) have convection plus a parameterized mixing prescription. They characterize the required spatial and temporal properties of the mixing. Lauer et al. (2018) evolve rotating models and allow for rotationally-induced mixing processes.

9 Microphysical inputs (e.g., opacities) often assume solar-scaled abundances 1 H 2 H 3 He 4 He 12 C 14 N 16 O. 56 Fe X Y Z

10 CNO enhancements occur on the H-rich AGB; there are convenient computational tools.

11 MESA can now use low-temperature, composition-dependent opacities T eff [K] log(l/l ) He Star R1 (X C = 0.012) R2 (X C = 0.081) log(t eff / K) M Reproducing Figure 3b of Weiss (1987)

12 Past MESA R CrB models have shown annoying numerical issues; hopefully, these are gone. Fig. from Zhang et al. (2014), see also Menon et al. (2013)

13 Mass loss recipes are an important ingredient. Generally, recent models use Bloecker-like winds with varying efficiencies, meaning the mass loss rate is 10 5 M yr 1. Menon et al. (2013), Zhang & Jeffery (2014), Lauer et al. (2018)

14 Mass loss recipes are an important ingredient. Generally, recent models use Bloecker-like winds with varying efficiencies, meaning the mass loss rate is 10 5 M yr 1. Menon et al. (2013), Zhang & Jeffery (2014), Lauer et al. (2018) For R CrB, v 300 km s 1, so the specific energy is 100x greater than in an AGB wind. What about the duty cycle? Clayton et al. (2003, 2013)

15 Accurate mass loss rates are required for accurate lifetime estimates. log(l/l ) t RCB = yr t RCB = yr t RCB = yr Ṁ = max[ṁmax, Bloecker(η = 0.02)] Ṁ max = M yr 1 Ṁ max = M yr 1 Ṁ max = M yr log(t eff / K) 4000

16 The CO WD cores don t grow significantly; the R CrB descendants don t have high masses. He WD Mass [M ] CO WD Mass [M ] Final core growth [M ] consistent with Zhang & Jeffery (2014)

18 Overview Models of RCrB Stars Models of their more massive cousins Summary

19 The merger of two CO WDs with a super-chandra total mass can collapse to an NS. 0.6 M 0.9 M CO + CO CO ONe! NS Si/Fe Nomoto & Iben; Saio & Nomoto (1985), JS et al. (2016)

20 It takes 20 kyr from merger to collapse; the models spend time in a cool giant phase. log(l/l ) ➍ 10 0 R 10 1 R 10 2 R 10 3 R 10 kyr ➌ 10 1 R 1 kyr ➋ ➊ 5 kyr log(t eff /K) JS et al. (2016)

21 The merger of He WD & ONe WD with a super-chandra total mass can collapse to an NS. 0.4 M 1.2 M He + ONe ONe AIC Brooks, JS et al. (2017b)

22 Massive R CrB analogs may be formed. Are there He-shell burning R CrB-like objects, but with more massive cores and hence noticeably higher luminosities? At these near-eddington luminosities, can you grow the core?

23 Overview Models of RCrB Stars Models of their more massive cousins Summary

24 Models of RCBs descended from double WD mergers are broadly consistent with observations, but we have to overcome uncertain mass loss rates uncertain mixing process More massive double WD mergers also make rare hydrogen-deficient objects that may sit in similar parts of the HR diagram. Even if they re 10% as common and live 10% as long, we may be able to find a few.

26 Opacity calculations are somewhat different than they were then log R = 4.5 log(κ/cm 2 g 1 ) R1 (WKM, Weiss 1987) R2 (WKM, Weiss 1987) R1 (OPAL, this work) R2 (OPAL, this work) log(t/k)

27 ... and low T is different too. log(κ/cm 2 g 1 ) R1 0 R2 log(ρ/g cm 3 )= 8 OPAL ÆSOPUS FA05 WKM log(t/k)

White Dwarf mergers: AM CVn, sdb and R CrB connections

White Dwarf mergers: AM CVn, sdb and R CrB connections Simon Jeffery Armagh Observatory many, many colleagues, but principally: Phil Hill, Uli Heber and Hideyuki Saio White Dwarf mergers: AM CVn, sdb and