SN 1994D in NGC 4526 Supernovae Type Ia Brightest Objects in Normal

9 MB ≈−19.5mag about 5 × 10 L⊙ Remarkable Standard Candles Baade 1938

∆MB ≈±0.5mag or ± 60% in L

Cosmology Projects A correlation between peak L and light curve shape

reduces MB to ±0.15mag or 15% in L Phillips A) SN 1998M z =0.63 B) SN 1998J z =0.83 C) SN 1997cj z =0.50 D) SN 1998I z =0.89 Type Ia Supernovae

Perlmutter, Physics Today (2003) 0.0001 26 Cosmology Project 24 High-Z Supernova Search 0.001 22 fainter Calan/Tololo 25 empty 0.01 Supernova Survey 0 20 mass 0.20.4 0.6 1.0 1 density 18 24 0.1 with vacuum energy Relative brightness 16 1 23 without vacuum energy 14 0.01 0.02 0.04 0.1

22 Accelerating Universe magnitude

21 Decelerating Universe

20 0.2 0.4 0.6 1.0 redshift

0.8 0.7 0.6 0.5 Scale of the Universe [relative to today's scale] Perlmutter et al. and Riess et al.

The Universe is accelerating – ΩΛ at z =1 SNe Ia are 30% fainter than if ΩΛ =0

(a factor of 2 fainter if ΩM =1, ΩΛ =0)

Nucleosynthesis SNe Ia are a major source of Fe – Fe is lost in neutron stars

Binary Star Evolution Progenitor evolution is convoluted Supernova Types Main Classification-spectral Type II – hydrogen lines (most common) Type I – no hydrogen lines Ib–noSilines Ic–noSiorHe Ia – prominent Si lines

Type II – deaths of massive stars M ≥ 8 M⊙

Degenerate processed core exceeds MCh – collapse to a neutron star

46 13 EB ≈ 3 × 10 J ≈ 10 L⊙ × 3 months – mostly in neutrinos – H-envelope is ejected

Type Ib/c – deaths of very massive stars M ≥ 40 M⊙ Core collapse after mass loss – Wolf–Rayet or binary stars Type Ia – thermonuclear explosions of CO white dwarfs Available nuclear energy can exceed the WD binding energy most reaches nuclear statistical equilibrium 56 and about 0.8 M⊙ of Ni is expelled 56Ni → 56Co → 56Fe observed 12 16 56 44 Energy – 1 M⊙ of 20% C + 80% O → Fe releases 1.8 × 10 J 56 56 43 Decay of 1 M⊙ of Ni → Fe releases 2 × 10 J of this 9 enough for 80days at 5 × 10 L⊙ White Dwarfs and the Cores of Giants Objects supported entirely by electron degeneracy pressure Heisenberg Uncertainty ∆x∆p ≥ ¯h/2 All available low-energy electron states occupied Reducing the volume squeezes electrons to higher momentum states Work must be done and a pressure is exerted

Maximum mass – Chandrasekhar – MCh =1.44 M⊙ Reached when electrons become relativistic

Nuclear ash accumulates in the cores of stars When all fuel is exhausted the core shrinks The star expands – erythrogigantism Energy per Nucleon Detonation by Mass Accretion

As MWD increases towards MCh collapse and heating can ignite fusion

CO WDs explode at about 1.38 M⊙ – Enuc >Ebind full processing to 56Ni

ONe WDs ignite too late – Ebind is not exceeded collapses to a NS – energy lost in neutrinos

He WDs ignite early when MWD ≈ 0.7M⊙ – Enuc ≫ Ebind degeneracy raised – no explosion? Mass Accretion requires a binary companion Central Ignition Ignition at the Outside Carbon Ignition Slow accretion Rapid accretion

Central cold fusion ignition Edge hot fusion ignition Burns to an ONe white dwarf How can the Companion get Close Enough?

The White Dwarf was the core of a Red Giant

Unstable Mass Transfer and Common Envelope Evolution Common Envelope Evolution

                 Cores Spiral Together

Envelope Lost Coalescence               Close Binary in Planetary Rapidly Spinning Giant

Gravitational 4 Radiation Magnetic Braking 10 yr

              Cataclysmic Variable Normal Giant The Standard Model At first sight cataclysmic variables are a tantalising possibility

Hot Spot

Accretion Disc Accretion Stream

White Dwarf Secondary Star

Cataclysmic Variable Star But nova explosions expel mass −7 −1 if M˙ ≤ 10 M⊙ yr and MCh is not reached

Nova Cygni 1992 – HST – after 467days Evolved Donor Stars (Symbiotics) - stellar evolution increases M˙

−7 −1 −7 if 10 ≤ M/M˙ ⊙ yr ≤ 3 × 10 hydrogen burns gently as it is accreted – Super Soft X-ray Sources but at higher rates a new giant envelope is formed – normal binaries do not spend enough time in between to grow the WD mass to MCh The Standard Model invokes a mechanism to keep −7 −1 M˙ < 3 × 10 M⊙ yr radiatively driven wind – Hachisu, Kato, Nomoto very efficient common envelope evolution – Nelemans, Tout But where is the hydrogen now? And where are the companions? The companion should be visible in remnants such as Tycho’s Ruiz-Lapuenta et al. 2004 but Kerzendorf et al. 2009 And where are the X-rays? The Super Soft Sources should be visible Gilfanov and Bogdan 2009 Merging CO + CO White Dwarfs The Standard Double Degenerate Model

M1 > M2 M2

Two close enough CO white dwarfs merge by gravitational radiation But if the mass accretion rate is too high carbon ignites non-degenerately near the outside and C is burnt to Ne throughout – can only end in AIC Merging CO + CO White Dwarfs Gravitationally driven mass transfer is too fast ˙ 1 ˙ Iben & Nomoto 1985 found M1 < 5 MEdd for central ignition

If mass ratio q = M1/M2 > 0.629 mass transfer is hydrodynamically unstable The white dwarf expands on mass transfer faster than the Roche lobe Positive Feedback Can this crush the centre before inside-out burning? Edge-Lit Detonations A close CO WD and He WD or naked helium star merge by GR

               He C/O           Like novae a He shell can ignite degenerately but now 0.1 − 0.2 M⊙ of He is required In 1–D models the thermonuclear runaway at the outside can set off the CO core ONe WDs with unburnt cores Some models of super-AGB stars in which C-burning begins non-degenerately off-centre seem to leave an unburnt CO core so • AIC could be avoided

• Easier to reach MCh because ONe WDs are more massive But this could be a numerical artefact in well-resolved models C burns close enough to the centre Why does the Peak Luminosity vary? It depends directly on the mass of 56Ni ejected

Current explosion models at 1.38 M⊙ show that typically the inner 0.2 M⊙ to 0.8 M⊙ of CO material burns to nuclear statistical equilibrium in a deflagration that ends at ρDD which is independent of composition.

In the inner 0.2 M⊙ weak interactions are important

Outside 0.8 M⊙ there is incomplete burning −→ Si So the mass of 56Ni depends mainly on the ratio of Z Z protons to neutrons N (or Ye = A ) Nuclear Statistical Equilibrium Neutron-Rich Material Dominated by 23Na from C burning and 22Ne from α-processed CNO Variation in CNO for 1 < Z < 3 can account for the variations in 3 Z⊙ peak luminosity – Timmes, Brown & Truran 2003 Note high z =⇒ low Z =⇒ brighter SNe TPAGB stars produce primary 22Ne Stancliffe & Tout 2006 The Phillips Relation Brighter SNe Ia have broader slower light curves So the light curve shape depends on 56Ni too?

1/2 3/4 −1/4 τLC ∝ κopt Mej Ek Arnett

κopt is the optical opacity of the ejecta – iron group isotopes – probably constant

Mej is the ejected mass – constant in standard models

Ek is the ejection energy – increases with C/O ratio

But C/O ratios are larger in low-Z progenitors! And the Correlation is in the wrong direction!

We must Understand the Progenitors first Conclusions

• Type Ia SNe are thermonuclear explosions of CO white dwarfs • The progenitors are otherwise unidentified – perhaps we haven’t even thought of the correct model yet? • There remains no straightforward explanation of the Phillips relation