Zorro and the nature of SNe Ia

Paolo A. Mazzali

Max-Planck Institut für Astrophysik, Garching

Astronomy Department and RESearch Centre for the Early Universe,

University of Tokyo

Istituto Naz. di Astrofisica, OATs

SN Rates, Florence 22.5.2008 2

Obs.Rel.1: The Phillips Relation (Absolute Magnitude - Decline Rate)

Phillips 1992

SN Rates, Florence 22.5.2008 3

O.R. 2: Bolometric Light Curves

Contardo et al. 2000

Lpeak - decline rate or LC shape

SN Rates, Florence 22.5.2008 4

SNe Ia @ maximum

“Branch normals”

SN Rates, Florence 22.5.2008 5

SNe Ia @ 20 days after maximum

“Branch normals”

SN Rates, Florence 22.5.2008 6

SNe Ia: late-time spectra

Mazzali et al. 1998

Δm15(B)

1.93

1.77

1.27

1.51

0.87

SN Rates, Florence 22.5.2008 7

O.R.3: Nebular line width - decline rate

after Mazzali et al. 1998

56Ni mass and distribution and decline rate (≡Luminosity) are related

SN Rates, Florence 22.5.2008 8

OR4: Velocity Gradients: an alternative SN Ia

classification Benetti et al. (2005):

• Classify SNeIa according to rate of change of post-maximum photospheric velocity of SiII 6355

3 SN groups: High Velocity GradientLow Velocity GradientFaint

• Groups separate out in v-ΔM15(B) plot

SN Rates, Florence 22.5.2008 9

II. Outer regions of the ejecta: HVFs

• Nearly all SNe show very High Velocity (~20000 km/s) absorption Features (HVF) in Ca II (some also in Si II)

Mazzali et al (2005)

SN Rates, Florence 22.5.2008 10

HVFs trace the outer explosion

What makes the HVFs?• Abundance enhancement

unlikely• Density enhancement

more reasonable (incl. H)• Ejection of blobs or CSM

interaction?• Blobs: line profiles

depend on orientation

SN Rates, Florence 22.5.2008 11

Distribution of HVFs

• Any model should reproduce the observed distribution of HVF wavelength (blob velocity) and strength (optical depth, covering factor)

• Single blob does not

SN Rates, Florence 22.5.2008 12

Need a few blobs or a thick torus

SN Rates, Florence 22.5.2008 13

Use blobs to fit spectra

SN Rates, Florence 22.5.2008 14

III. Abundance Stratification

Model spectral sequence to derive composition layering: explore the depth of the star

How did the star burn?

Stehle et al 2005

SN Rates, Florence 22.5.2008 15

Modelling late-time spectra

Full view of inner ejecta (56Ni zone)

Monte Carlo LC code + NLTE nebular code

• No radiative transfer• Estimate masses

SN Rates, Florence 22.5.2008 16

Result: Composition in a typical SN Ia

• SN 2002bo

m15(B)=1.15

(average)

• Elements more mixed than in typical 1D models

SN Rates, Florence 22.5.2008 17

Test: Modelling Light Curve

Monte Carlo code• Use W7 density• Composition from

tomography

• 56Ni ~0.50M Model LC matches

data very well

SN Rates, Florence 22.5.2008 18

IV. The Global View

• Late time spectra suggest M(56Ni) Δm15(B) [ M(Bol)] v(Fe)

SN Rates, Florence 22.5.2008 19

Role of 54Fe, 58Ni

• Stable Fe group isotopes radiate but do not heat• Some anticorrelation between 56Ni and (54Fe, 58Ni)• Very good correlation beween Σ(NSE) and Δm15(B)

SN Rates, Florence 22.5.2008 20

Composition Layering

• Outer extent of NSE zone varies ( Δm15(B), Lum)• Inner extent of IME matches outer extent of NSE• Outer extent of IME ~ const.

SN Rates, Florence 22.5.2008 21

Putting it all together: “ orro” diagramA basic property of SNe Ia

Mass burned ~ constant

Progenitor mass also probably constant: MCh

KE ~ const

What does it all mean?• Delayed detonation?• Multi-spot ignited deflagration?• Other possibilities….?

Stable 54Fe, 58Ni

56Ni

IME

Mazzali et al. (2006), Science

SN Rates, Florence 22.5.2008 22

The Role of Fe-group, IME on LC

• 56Ni: light, opacity, KE

• 54Fe, 58Ni: opacity, KE• IME: KE, some opacity• CO (if any): little opacity

SN Rates, Florence 22.5.2008 23

V. Explaining the Phillips’ Relation

Models with increasing 56Ni content reproduce Phillips’ Relation

(Mazzali et al. 2001)

SN Rates, Florence 22.5.2008 24

Spectra also reproduced

Mazzali et al. 2001

-7days Max

SN Rates, Florence 22.5.2008 25

Using Zorro to reconstruct Phillips’ Rel’n

• Use composition to compute LC parameters

•

•

•

•

• Derive Phillips Relation

SN Rates, Florence 22.5.2008 26

Using Zorro to reconstruct Phillips’ Rel’n

• Use composition to compute LC parameters • Derive Phillips Relation

SN Rates, Florence 22.5.2008 27

Effect of stable NSE (54Fe, 58Ni)

• Amount of 54Fe, 58Ni may be a fn. of WD metallicity (Timmes et al. 2003)• Ratio 56Ni/NSE affects light, not KE or opacity LCs w/ different peak brightness, same decline rate May cause spread about the Phillips’ Relation. (Mazzali & Podsiadlowski

2006)

SN Rates, Florence 22.5.2008 28

Conclusions• There is some regularity among SNe Ia (surprise

surprise…)

• Ejecta reflect stratified composition of models

• Total mass burned may be constant

• 56Ni determines luminosity (not new)

• Total NSE determines LC shape

• Degree of neutronization can give rise to dispersion in luminosity-decline rate relation

• Z evolution may confuse cosmological parameters