Modelling thermonuclear supernovae Stuart Sim Mattia Bulla, Mark - PowerPoint PPT Presentation

Modelling thermonuclear supernovae Stuart Sim Mattia Bulla, Mark Magee, Markus Kromer, Fritz Röpke, Ashley Ruiter, Ivo Seitenzahl, Wolfgang Hillebrandt

Supernovae in astrophysics • Explosive death of star – dramatic end point of stellar evolution • Nuclear burning in SNe makes the heavy elements • Inject energy, momentum and metals; can affect galaxy evolution • Type Ia “ Standardizable candles ”, probes of expansion history of the Universe • Challenge our understanding of physics – Turbulence and hydrodynamics – Combustion and flame physics – Nuclear physics – Radiative transfer SN1994D in NGC 4526 NASA/HST Melbourne 2017

Overview • Thermonuclear supernovae – Reminder of basic picture for Type Ia supernovae • Explosion models – Chandrasekhar mass explosions • Pure deflagrations and SNe Iax (Magee) – Sub-Chandrasekhar mass models Melbourne 2017

Established picture for a thermonuclear supernova explosion Energy release by flame unbinds star CO WD Ignition, Flame ~2 sec. birth of flame propagation IMEs 56 Ni Peak of optical display as energy 56 Ni from 56 Ni -> 56 Co radioactivity -> 56 Fe Ejecta expand; Radioactive escapes homologous by ~100 sec. decays reheat ~20 days ejecta Melbourne 2017

Supernovae Ia ~0.7 M ¤ of 56 Ni is typical Melbourne 2017

Supernovae Ia Made of Si, S and Fe H and He not detected Velocities measured from lines v ~ 15 , 000 km/s Melbourne 2017

Supernovae Ia: diversity Diversity in SNe Ia and related transients (figure from Taubenberger 2017) (rates from Li et al. 2011) Melbourne 2017

Supernovae Ia Li et al. 2011 (Lick Obs. SN. Search) Melbourne 2017

Thermonuclear Supernovae Many unanswered questions remain: • How did the system evolve to ignition? • Progenitor channel debate (“single vs double degenerate”) • What are the properties of the exploding star and how do these affect what we see? • Mass of WD, composition, immediate environment? • Where does the flame ignite and how does it propagate (deflagration, detonation)? Melbourne 2017

SN Ia Flame Basics • Instantaneously narrow region in which nuclear reactions are taking place • Thermonuclear flame propagation modes • Deflagration ( sub-sonic) • Detonation ( supersonic) • Flame generates energy (nuclear burning) • Eventually unbinds star Melbourne 2017

Synthetic explosions: testing models by comparing to data Spectra / photometry Explosion Hydrodynamical for SN Ia conditions simulations and final state Radiative transfer simulations Ignition conditions Parameterised Composition structure Theories for progenitor(s) Nucleosynthesis calculations Melbourne 2017

Testing explosion scenarios Melbourne 2017

Explosion scenarios Best known paradigm: (Near-)Chandrasekhar-mass single-degenerate scenario – WD in binary with H-rich star (main-sequence or giant) – Mass-transfer – Mass is retained (avoid net mass- loss in nova explosions) – H burned to He then C/O – WD grows in mass: central density and temperature rise Melbourne 2017

Explosion scenarios Best known paradigm: (Near-)Chandrasekhar-mass single-degenerate scenario Explosion mechanism: – WD heated by C burning during ~1000 yr “simmering” phase (Kuhlen et al. 06, Zingale et al. 06, 11) – Thermonuclear runaway occurs – Deflagration born (prompt detonation is no good for Ch mass) – Proceeds as pure deflagration? Melbourne 2017

Deflagration models 3D simulation: Kromer+ 13

Deflagration models 1600C 300C 1 . 6 1600 200 1 . 4 M tot 40 100L ejected mass ( M � ) 100H 1 . 2 100 150 1 . 0 20 0 . 8 0 . 6 M IGE 10 5 0 . 4 M 56 Ni 3 0 . 2 1 M IME 0 . 0 0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0 1 . 2 1 . 4 E nuc ( 10 51 erg ) Sequence of models: Fink+14. – roughly 1 order of magnitude in 56Ni mass. Bound remnant found in some cases, in agreement with Jordan+12.

Supernovae Ia: diversity Multiple sub-classes of SNe Ia / related transients (figure from Taubenberger 2017) Melbourne 2017

Supernovae Ia: diversity Faint with weak IMEs Bright with early IGE lines Also a sample of very bright cases e.g. 2003fg (Howell+); 2007if (Scalzo+, Yuan+); 2009dc (Yamanaka+)... Faint and fast And a handful of faint, slowly evolving cases PTF10ops (Maguire+); There are now multiple sub-classes of SNe Ia 2010lp (Pignata+) (from Li et al. 2011) Melbourne 2017

Pure deflagration: 02cx-like SNe? Foley+ 13 Pure deflagration models: − 18 Suggested connection to peculiar (faint) Ia’s: − 17 Absolute V Magnitude Branch+04, Jha+06, Phillips+07 − 16 Now evidence that 02cx-like class (“SNe Iax”; Foley+09,13) − 15 • is large and diverse • 11ay range of ejecta mass − 14 12Z 08A (“failed” deflagrations) 05hk 08ge 08ae − 13 02cx 03gq 05cc 09J 08ha − 12 0 20 40 60 Rest − Frame Days Relative to V Maximum

Supernovae Ia: a progenitor? First plausible detection of a Ia progenitor: (from McCully et al. 2014) Melbourne 2017

Deflagration models Spectra of 05hk from Phillips+07

Deflagration models Spectra of 05hk from Phillips+07 - pretty good match to model (Kromer+ 13)

Deflagration models N5 compared to 05hk – late times ?

Deflagration models Comparison Days since maximum -20 0 20 40 60 80 100 120 140 extended to fainter K - 2 example: H - 1.5 15 J - 1 i - 0.5 SN2015H 16 r Apparent Magnitude V + 0.5 Magee+2016 g + 1 17 18 19 20 57040 57060 57080 57100 57120 57140 57160 57180 57200 57220 MJD

Deflagration models -20 Comparison Normal Ia extended to fainter -19 11hyh 09ego example: 09ku 12Z N5def 11ay Peak Absolute Magnitude 08A -18 05hk PS15csd 08ae N3def SN2015H 02cx 03gq 15H -17 05cc Magee+2016 N1def 04cs 09eoi -16 07qd 09J -15 -14 SNe Iax r R 08ha 10ae Normal SNe Ia Model 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Decline Rate

Deflagration models -19 g V N5def SN2015H -18 N3def Magee+2016 N1def -17 -16 Absolute Magnitude -15 -14 r i -18 -17 -16 -15 -14 5 15 25 35 45 5 15 25 35 45 Days since explosion

Deflagration models SN2015H Magee+2016 (22 days)

Deflagration models: summary Strengths: • Full star, multi-D deflagration simulations – explosions occur • Star not (always) full disrupted • Synthetic spectra are fair matches to observed SN Iax class • Peak luminosities (and colours) match fairly well

Deflagration models: summary Strengths: • Full star, multi-D deflagration simulations – explosions occur • Star not (always) full disrupted • Synthetic spectra are fair matches to observed SN Iax class • Peak luminosities (and colours) match fairly well Open issues: • Light curve timescales too fast in models: need more ejecta mass? • Support for alternative models (e.g. PDD models; Stritzinger et al. 2015) • Major challenge: late times • Major challenge: very faint objects (08ha needs only ~3x10 -3 M sun of 56 Ni)

Deflagration model: late evolution Comparison of bolometric light curve out to late times (Kromer+13) Energy deposited in the “bound” remnant? Jha+06, Sahu+08, Foley+16, Shen+17

Deflagration models -20 Comparison Normal Ia extended to fainter -19 11hyh 09ego example: 09ku 12Z N5def 11ay Peak Absolute Magnitude 08A -18 05hk PS15csd 08ae N3def 02cx 03gq 15H -17 05cc N1def 04cs 09eoi -16 07qd 09J -15 -14 SNe Iax r R 08ha 10ae Normal SNe Ia Model 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Decline Rate

Deflagration models Conclusion: (Jordan et al. 2012; Kromer et al. 2013, Fink et al. 2014, Kromer et al. 2015, Magee et al. 2016) Near-Chandrasekhar mass WD deflagrations may work well for the 2002cx-like SNe Ia …still multiple loose ends …what are the “normal” ones?!