type ia supernovaeicc.ub.edu/webs/iccub_winter_meeting/sites/icc.ub... · type ia supernovae: the...

Carles Badenes University of Pittsburgh / PITT PACC / ICCUB

Type Ia Supernovae:The search for the Progenitors

ICCUB Winter MeetingFebruary 1, 2018

Research Group● Postdocs: Brett Andrews (2014-2017).

● Graduate Students: Matt Schell, Thomas Hettinger (PhD 2015, MSU), Sumit Sarbadhicary (PhD 2018), Héctor Martínez-Rodríguez, Christine Mazzola.

● Collaborators: Hiroya Yamaguchi (NASA), Eduardo Bravo (UPC), Tony Piro (OCIW), Todd Thompson (OSU).

Carles BadenesICCUB 01/02/18

Papers up to 2017

Today: SNRs and SN Ia progenitorsLate Feb (20?): Binaries in surveys

Carles BadenesICCUB 01/02/18Outline

The SN Ia Progenitor Problem

Supernova Remnants

Neutron-rich Isotopes in Type Ia SNRs

The SN Ia Progenitor Problem

SN 1994D(P. Challis)

Carles BadenesICCUB 01/02/18Supernovae (SNe)

Wavelength [A]

Rel

ativ

e Fl

ux

SNe are rare optical transients (t ~ months) with peak magnitudes that rival

their host galaxies

● Types (optical spectra):● Type I: no H (Ia: Si, Ib: He,

Ic: weak Si, no He).● Type II: Strong H.

● Core collapse SNe (II, Ib, Ic): massive stars (M ≥ 8M

⊙),

several progenitors identified.● Thermonuclear SNe (Ia).

D. Kasen

Carles BadenesICCUB 01/02/18Type Ia SNe and Cosmology

SupernovaCosmology Project

● Explains basics: no H, energy budget, rate of light curve decay.

● Key details remain obscure: explosion mechanism, progenitor systems.

● Strikingly uniform light curves LC width / luminosity relation ⇒(56Ni mass) ⇒ Cosmology.

SN Ia are thermonuclear explosions of C+O white

dwarfs prompted by accretion in a binary system

Carles BadenesICCUB 01/02/18Type Ia SNe and Cosmology

The Nobel Prize in Physics 2011 was divided, one half awarded to Saul Perlmutter, the other half jointly to Brian P. Schmidt and Adam G. Riess "for the discovery of the accelerating expansion of the Universe through observations of distant Type Ia supernovae".

Carles BadenesICCUB 01/02/18How to Explode a White Dwarf

DD System

SD System

Carles BadenesICCUB 01/02/18Single vs. Double Degenerate SN Ia

Single Degenerate (SD): ● WD+non-degenerate star.● Slow accretion mass growth ⇒ ⇒

explosion near MCh [Hachisu+ 96].

● Some CSM expected (accretion cannot be 100% efficient) [Han & Podsiadlowski 04].

Double Degenerate (DD): ● WD+WD.● GW emission merging/collision ⇒ ⇒

explosion, not necessarily near MCh [Iben & Tutukov 84, Webbink 84, Sim+ 10, van Kerkwijk+ 10, Thompson 11].

● Not much CSM expected. DD System

SD System

Carles BadenesICCUB 01/02/18Single vs. Double Degenerate SN Ia

Carles BadenesICCUB 01/02/18SN Ia Progenitor Scenarios

Naïve expectations:

Accretors (SD) should explode close to MCh and have some CSM

Mergers and collisions (DD) might or might not explode close to MCh and might or might not have CSM

References: Wang & Hang 12; Maoz+14; Hachisu+ 96; Iben & Tutukov 84; Webbink 84; Kashi & Soker 11

Carles BadenesICCUB 01/02/18Does it Matter?

● Oh yes. Evolution of SN Ia progenitors with redshift would introduce biases in 'precision cosmology' measurements.

● Systematics cannot be beat with statistics.

● Crucial for WFIRST and LSST, but cannot be solved by them alone. There is no substitute for actual understanding!

WFIRST team (D. Scolnic, R. Foley)

Carles BadenesICCUB 01/02/18Shouldn't we know by now?

● Most important open question in stellar astrophysics (2010 decadal survey).

● Open Supernova Catalog [Guillochon+ 16] has >3,000 Type Ia SNe with spectra. In some cases, the data is excellent (see right for SN2011fe).

● Sub-Ch SN Ia might be 1.2 M ⊙(vs. 1.4

M ⊙for MCh) hard to distinguish with ⇒

optical spectroscopy.

● Optical SN spectra insensitive to CSM interaction.

● Data dimensionality more important that data quantity ⇒ look elsewhere!

Supernova Remnants

SN 1994D(P. Challis)

Tycho SNR (SN 1572)Chandra

Supernova Remnants (SNRs) Carles BadenesICCUB 01/02/18

Forward shock

Reverse shock

Shockedejecta

● SNR = SN + ambient medium (AM) interaction.

● Collisionless shocks, v~5000 km/s ⇒ X-ray temperatures (~1 keV).

● Forward shock moves out (ISM/CSM), reverse shock moves in (SN ejecta).

● X-ray spectrum has prominent lines from shocked SN ejecta.

Shocked AM

Supernova Remnants (SNRs) Carles BadenesICCUB 01/02/18

● SNR = SN + ambient medium (AM) interaction.

● Collisionless shocks, v~5000 km/s ⇒ X-ray temperatures (~1 keV).

● Forward shock moves out (ISM/CSM), reverse shock moves in (SN ejecta).

● X-ray spectrum has prominent lines from shocked SN ejecta.

Tycho SNR(Chandra)

AM Interaction in SNRs Carles BadenesICCUB 01/02/18

● AM structure progenitor mass-loss. Modified circumstellar ⇒medium (CSM) vs. undisturbed interstellar medium (ISM)● SNe Follow-up (radio/X-ray) probes to ~ 100 AU.⇒● SNRs spatial (and temporal) scales relevant for stellar ⇒evolution of SN progenitors (t τ≲ KH). Probe dynamical interaction!

AM Interaction in SNRs

● X-ray spectra ⇒ constrain AM structure. NEI plasma: ionization timescale (net) [Badenes+ 07]. ● High net high centroid ⇒energy and line flux.

Badenes+ 07

Silicon net 5x109 to 1012 cm-3 s

H-likeHe-like


AM Interaction in SNRs: Fe K

● Use Fe Kα line blend at ~6.5 keV as an AM density diagnostic.

● Most SNe (Ia and CC) eject some Fe ⇒innermost layers.

● Large net required to fully ionize Fe ⇒ large dynamic range in ρAM.

● Details: Yamaguchi, CB+ 14b.

● Need high effective area at 6.5 keV: Suzaku.


Badenes+ 08a

Yamaguchi+ 14a

AM Interaction in SNRs: Fe K

● Use Fe Kα line blend at ~6.5 keV as an AM density diagnostic.

● Most SNe (Ia and CC) eject some Fe ⇒innermost layers.

● Large net required to fully ionize Fe ⇒ large dynamic range in ρAM.

● Details: Yamaguchi, CB+ 14b.

● Need high effective area at 6.5 keV: Suzaku.


Suzaku X-ray satellite (2005-2015)

Models vs. Data

● Suzaku data for 23 SNRs + Chandra, XMM [Yamaguchi+ 14, Borkowski+ 13, Maggi+ 16].

● Evaluate stellar evolution + explosion with SNR observations.

● Models are required to interpret these data.

● Uniform AM, MCh ejecta can explain (most) Ia SNRs.


DEM L71(XMM)

Type Ia SNR Models

● Type Ia SNR models: MCh ejecta + uniform AM evolved to 5000 yr [Badenes+ 03,05,06,08a].

● DDT ejecta models (dim, normal, bright SN Ia) crude (but ⇒effective) diagnostic of SN Ia brightness!

● Also PDD models more compact ⇒

ejecta.


What is going on?

Most Type Ia SNRs show no clear signs of CSM

interaction


● Different dynamics for CC and Ia SNRs: several M

⊙

of CSM vs. much less, maybe no CSM at all.

● Most Ia SNRs compatible with ISM interaction. Slow isotropic outflows active at explosion ruled out. ● Kepler, N103B likely have some CSM [Patnaude+ 12, Burkey+ 12, Chiotellis+ 12, Williams+ 14].

● RCW 86 is a cavity explosion [Badenes+ 07, Williams+ 11, Broersen+ 14].

A Step Back Carles BadenesICCUB 01/02/18

● SN Ia AM density estimates from radio/X-ray SNe (~10d, ~0.01 pc) and SNRs (~500 yr, ~several pc) are consistent with the warm phase of the ISM [Badenes+ 07, Chomiuk+ 12, 16, Perez-Torres+ 14, Raymond+ 07, Slane+ 14, Borkowski+ 14]. 'clean' mergers (DD)?⇒● Mild CSM interaction allowed, maybe small (~0.5 pc) cavities [Patnaude+ 12, Slane+ 14], but not large ones (except for RCW86!).

What does it all mean?

● Fe K line CC/Ia + quantitative test for progenitor evolution. ⇒Most Ia SNRs compatible with little or no CSM mergers?⇒

● SNR 3C397 prominent Mn and Ni emission ⇒ M⇒ Ch progenitor accretor?⇒

● Other measurements show a preference for mergers (missing companions, DTD, WD merger rate), but suggest a non-negligible fraction of MCh explosions (light curves, solar Mn abundance).


SN Ia in star-forming galaxies probably come from a mixture of mergers and accretors

Neutron-rich Isotopes in Type Ia SNRs

Supernova Nucleosynthesis Carles BadenesICCUB 01/02/18

● All elements heavier than O are made in SNe.

● Metallicity (Z) mass ⇒fraction of elements heavier than He (for the Sun, Z

⊙

=0.014).

● Z increases as more SNe pollute the ISM rough ⇒proxy for stellar age. Often expressed logarithmically as [Fe/H].

● SN Ia make most Fe-peak nuclei. Main product is 56Ni which decays to 56Fe.

Solar metallicity

SN Ia Nucleosynthesis● Burning regimes in SN Ia: Explosive O burning, exp. Si burning, NSE, n-NSE Si, S, Ar, ⇒Ca, Fe [Thielemann+ 86].● How are n-rich isotopes (55Mn,60Ni, ...) produced? CO WDs have no neutron excess! Whence do neutrons come from?

● Neutron excess η:


Thielemann+ 86

12C η = 0⇒16O η = 0⇒

SN Ia Neutronization Carles BadenesICCUB 01/02/18

12C η = 0⇒16O η = 0⇒22Ne ⇒ η ⇔ Z

Neutron excess in Type Ia SNe:

● Progenitor metallicity: CNO cycle bottleneck is 14N(α,γ) ⇒hydrostatic He-burning ⇒ 22Ne. Then, η = 0.101xZ [Timmes+ 03, Badenes+ 08]. Works for all progenitors.

● n-NSE: During explosion, e-captures in NSE at high densities (ρ 10≳ 8 g cm-3) η (Fe-peak ⇒yields). Requires MWD=MCh !!

● For simplicity, I am ignoring a third way, C-simmering [Bildsten & Piro 08, Martínez-Rodríguez + 16]

Martinez-Rodriguez+ 16

n-NSE

Neutron-Rich Isotopes in SN Ia

● MCh DDT explosions (standard SN Ia models) [Khokhlov 91]. One parameter (ρtr) ⇒ 56Ni yield (SN Ia brightness).● Sub-Ch explosions also viable [Sim+ 10]. One parameter (MWD) ⇒56Ni yield.● Sub-Ch models do not reach n-NSE ⇒ smaller yield of neutronized species (Mn, Ni).● Tentative association:

● MCh DDT SD High Mn, Ni⇔ ⇔

● Sub-Ch DD⇔ Low Mn, Ni⇔


Neutron-Rich Isotopes in SN Ia Carles BadenesICCUB 01/02/18

● MCh DDT explosions (standard SN Ia models) [Khokhlov 91]. One parameter (ρtr) ⇒ 56Ni yield (SN Ia brightness).● Sub-Ch explosions also viable [Sim+ 10]. One parameter (MWD) ⇒56Ni yield.● Sub-Ch models do not reach n-NSE ⇒ smaller yield of neutronized species (Mn, Ni).● Tentative association:

● MCh DDT SD High Mn, Ni⇔ ⇔

● Sub-Ch DD⇔ Low Mn, Ni⇔

● Yield of neutronized species: n-NSE + progenitor metallicity [Timmes+ 03, Badenes+ 08b].

● Mn and Ni are hard to observe in the optical [Maeda+ 10, Seitenzahl+ 13].

● Diagnostic mass ratios for SNRs: MNi/MFe and MMn/MFe


Z

Neutron-Rich Isotopes in SN Ia

MNi/MFe and MMn/MFe can discriminate Ch and Sub-

Ch SN Ia progenitors

Carles BadenesICCUB 01/02/18Neutron-Rich Isotopes in SN Ia

● Yield of neutronized species: n-NSE + progenitor metallicity [Timmes+ 03, Badenes+ 08b].

● Mn and Ni are hard to observe in the optical [Maeda+ 10, Seitenzahl+ 13].

● Diagnostic mass ratios for SNRs: MNi/MFe and MMn/MFe

MNi/MFe and MMn/MFe can discriminate Ch and Sub-

Ch SN Ia progenitors

SNR 3C397 Carles BadenesICCUB 01/02/18

● 3C397 is an evolved Type Ia SNR at D~10 kpc [Safi-Harb+ 05]. ● Consistent dynamical model (IR+X-ray) ⇒ RS has thermalized all the SN ejecta.● Extraordinary X-ray spectrum! Very strong Ni and Mn emission.


● Model line emission with updated atomic data (AtomDB, Foster+) ⇒MNi/MFe~0.2; MMn/MFe~0.03.

● Sub-Ch models do not work, or require unreasonable progenitor metallicities (>5Z

⊙).

● MNi/MFe and MMn/MFe

require n-NSE material ⇒Chandrasekhar-mass progenitor. ● Details: Yamaguchi, CB + 15 ApJ 801, L31.


SD System

● A VERY metal-rich sub-Ch WD progenitor is extremely unlikely. Stars of the required metallicity (~5 Z

⊙) do not exist in the Milky

Way (cannot invoke lake Wobegone effect).

● Evidence for some MCh SN Ia progenitors is now growing [Seitenzahl+ 13, Scalzo+ 14].

SNR 3C397 had an MCh progenitor, probably SD Solar metallicity

What does it all mean?

● Fe K line CC/Ia + quantitative test for progenitor evolution. ⇒Most Ia SNRs compatible with little or no CSM mergers?⇒

● SNR 3C397 prominent Mn and Ni emission ⇒ M⇒ Ch progenitor accretor?⇒

● Other measurements show a preference for mergers (missing companions, DTD, WD merger rate), but suggest a non-negligible fraction of MCh explosions (light curves, solar Mn abundance).



And now what?

● Make a census of masses in Type Ia SNRs extend ⇒measurements of neutron excess to other SNRs, include the effect of C simmering ⇒ Martínez-Rodríguez thesis.● Find out exactly how much CSM can be allowed around accretors by the Fe K centroid measurements in SNRs ⇒ Schell thesis. ● Determine the contribution of different channels to the SN Ia rate

Delay Time Distribution ⇒ ⇒ Sarbadhicary thesis.



Back to the Future

Carles BadenesICCUB 01/02/18The future...

“I propose great things for inspection and contemplation by every explorer of Nature. Great, I say, because of their excellence,

because of their newness, unheard of through the ages, and

also because of the instrument with the benefit of which they

make themselves manifest to our sight..”

Carles BadenesICCUB 01/02/18The future... from the past

“I propose great things for inspection and contemplation by every explorer of Nature. Great, I say, because of their excellence,

because of their newness, unheard of through the ages, and

also because of the instrument with the benefit of which they

make themselves manifest to our sight..”

– Galileo Galilei, Sidereus Nuncius (1610)

Carles BadenesICCUB 01/02/18Future Plans

● Current and future surveys are opening new fields for scientific discovery.

● Rich landscape, full of possibilities, that rewards lateral thinking, diversity, and opportunism.

● One example: time domain in large spectroscopic surveys statistical ⇒analysis of binary systems.

Carles BadenesICCUB 01/02/18From Badenes & Maoz 2012

Carles BadenesICCUB 01/02/18Future Plans

Carles BadenesICCUB 01/02/18Diversity and Opportunism

● Sarbadhicary, S., Badenes, C., Chomiuk, L., Caprioli, D., & Huizenga, D. (2016) Supernova Remnants in the Local Group I: A model for the radio luminosity function and visibility times of supernova remnants. MNRAS, in press arXiv:160504923S● García-Berro, E., Badenes, C., Aznar-Siguán, G. & Lorén-Aguilar, P. (2016) White dwarf dynamical interactions and fast optical transients. MNRAS, submitted.● Martínez-Rodríguez, H., Piro, A. L., Schwab, J., & Badenes, C. (2016). Neutronization During Carbon Simmering In Type Ia Supernova Progenitors. ApJ, 825, 57● Hettinger, T., Badenes, C., Strader, J., Bickerton, S. J., and Beers, T. C. (2015). Statistical Time-resolved Spectroscopy: A Higher Fraction of Short-period Binaries for Metal-rich F- type Dwarfs in SDSS. ApJ, 806:L2● Yamaguchi, H., Badenes, C., et al.. (2015). A Chandrasekhar Mass Progenitor for the Type Ia Supernova Remnant 3C 397 from the Enhanced Abundances of Nickel and Manganese. ApJ, 801:L31● Badenes, C., Maoz, D., & Ciardullo, R. (2015). The Progenitors and Lifetimes of Planetary Nebulae. ApJ, 804:L25● Yamaguchi, H., Badenes, C., et al. (2014). Discriminating the Progenitor Type of Supernova Remnants with Iron K-shell Emission. ApJ, 785:L27● Park, S., Badenes, C., Mori, K., Kaida, R., Bravo, E., Schenck, A., Eriksen, K. A., Hughes, J. P., Slane, P. O., Burrows, D. N., & Lee, J.-J. (2013). A Super-solar Metallicity for the Progenitor of Kepler’s Supernova. ApJ, 767:L10● Badenes, C., et al. (2013). SDSS 1355+0856: a detached white dwarf + M star binary in the period gap discovered by the SWARMS survey. MNRAS, 429:3596–3603● Patnaude, D. J., Badenes, C., Park, S., and Laming, J. M. (2012). The Origin of Kepler’s Supernova Remnant. ApJ, 756:6● Maoz, D., Badenes, C., & Bickerton, S. J. (2012). Characterizing the Galactic White Dwarf Binary Population with Sparsely Sampled Radial Velocity Data. ApJ, 751:143● Badenes, C. & Maoz, D. (2012). The Merger Rate of Binary White Dwarfs in the Galactic Disk. ApJ, 749:L11● García-Senz, D., Badenes, C., & Serichol, N. (2012). Is There a Hidden Hole in Type Ia Supernova Remnants? ApJ, 745:75● Bravo, E. & Badenes, C. (2011). Is the metallicity of their host galaxies a good measure of the metallicity of Type Ia supernovae? MNRAS, 414:1592–1606

1st & 2nd author papers + GS papers since 2011Press releases:

Steps Forward

● Expand the model grid for Type Ia SNRs: CSM interaction, sub-Chandra explosions (Matt Schell's thesis).

● Improve the model physics: CR-modified dynamics [Lee+ 14].

● CC SNR models. Evaluate SN and progenitor models at the same time [Patnaude+15].

Patnaude+ 15


Beyond Ni and Mn Carles BadenesICCUB 01/02/18

● Progenitor metallicity (η) for Tycho, Kepler, and 3C397 from Mn and Ni lines: solar or super-solar [Badenes +08, Park+ 13, Yamaguchi+ 15].

● Ca yield also depends on η [De +14, Miles+ 16], unaffected by n-NSE!!!

● Measure η in more SNRs, including those in the LMC! Work by Héctor Martínez-Rodríguez in prep.

● For Tycho, Kepler, and 3C397, we recover the high values of η.

Normal and energetic SN Ia models (subenergetic ruled out by Fe K centroids [Yamaguchi+ 14])

Beyond Ni and Mn Carles BadenesICCUB 01/02/18

● Progenitor metallicity (η) for Tycho, Kepler, and 3C397 from Mn and Ni lines: solar or super-solar [Badenes +08, Park+ 13, Yamaguchi+ 15].

● Ca yield also depends on η [De +14, Miles+ 16], unaffected by n-NSE!!!

● Measure η in more SNRs, including those in the LMC! Work by Héctor Martínez-Rodríguez in prep.

● For Tycho, Kepler, and 3C397, we recover the high values of η.

Normal and energetic SN Ia models (subenergetic ruled out by Fe K centroids [Yamaguchi+ 14])

Very high η for 3C397 & N103B!!!

Neutronization from metallicity? Carles BadenesICCUB 01/02/18

● Lake Wobegon effect need to compare to Milky Way and LMC ⇒metallicity distributions [Hayden+ 15, Piatti & Gleiser 12]. ● There must be some extraneous (non-metallicity) source of neutrons simmering? ⇒ Requires MWD=MCh!

Yes, but... Carles BadenesICCUB 01/02/18

● Recent study of C simmering in accreting WDs with MESA limits η from simmering to ~1/3Z

⊙ [Martínez-Rodríguez + 16].

● Not enough to reproduce observed η, but models are simple. Longer simmering timescale? Additional cooling from axions? ● Simmering remains the only viable candidate for the origin of the neutron excess seen in the Ca/S measurements of SNRs! ⇒progenitors must accrete slowly before they explode!

Track of WD center

Yes, but... Carles BadenesICCUB 01/02/18

Type Ia SNRs and cavities

RCW 86

RCW 86

Badenes+ 07RCW 86

● Radii and net of Type Ia SNRs with known ages are consistent with uniform ambient medium interaction [Badenes+ 07].

● 'Accretion winds' in SD progenitor models [Hachisu+ 96] excavate large cavities [Koo & McKee 92] that lead to large SNR radii and low net.

Carles BadenesCfA 02/19/15

More on RCW 86

● RCW 86 is large (~25 pc), with well defined borders, low net, bright Fe, and no compact remnant [Williams+ 11].

● IF SNR of SN 185 AD ⇒ cavity explosion [Vink+ 97].

● IF Ia SNR ⇒ fast, sustained outflow from the progenitor ⇒ SD [Badenes+ 07, Williams +11].

● A light echo or detailed HD+NEI models would be very nice!

Williams+ 11

RCW 86


Other cavity Ia SNRs?

● RCW 86 might not be the only example of Type Ia SN in a cavity.

● DEM L238 and DEM L249, two middle-aged SNRs in the LMC have Fe-rich spectra and low net.

● IF Type Ia SNRs, they might also be cavity explosions [Borkowski+ 06].

● Beware: typing SNRs older than a few thousand years is difficult, and so is modeling their dynamic evolution!

DEM L238 DEM L249

Borkowski+ 06


CSM in CC SNRs

● In more evolved SNRs like G292.0+1.8, forward shock morphology can constrain ejecta and CSM density profiles ⇒ CC SN progenitor [Lee+ 10]. Lee+ 10


CSM Interaction: Kepler SNR

60

● Kepler is unique among Type Ia SNRs in that it shows clear signs of a non-uniform AM in the NW: brighter X-ray emission, larger net, lower expansion parameters, optical N-rich emission [Blair+ 91, Reynolds+ 07, Vink 08].

● Well above Galactic plane ⇒ CSM from a mass-losing progenitor. A popular model posits a large relative motion wrt to the local ISM ⇒ bow shock structure overrun by SN ejecta [Bandiera 87, Borkowski+ 92, 94].

Carles Badenes Garching September 13, 2012

Vink 08

Some Nuclear Physics Carles BadenesICCUB 01/02/18


62Carles Badenes Garching September 13, 2012

● Morphology (radius and N/S asymmetry) and kinematics (expansion parameters) can be reproduced by a symbiotic model (AGB wind ~20 km/s, moving at 250 km/s wrt ISM) [Chiotellis+ 12].

● However, this requires a subenergetic SN explosion (E~2x1050 erg).

Chiotellis+ 12


63

● HD+NEI models in the S, where the ejecta should be interacting with the pristine CSM from the progenitor ⇒ constrain both M56Ni and pre-SN dM/dt [Patnaude+ 12].


Patnaude+ 12


64

● HD+NEI models rule out a standard ρ ∝ r-2 CSM! (allowed by HD [Chiotellis+ 12]).● Small cavity + wind works [Wood-Vasey & Sokoloski 06], but so does a uniform AM.● In any case, Kepler must have been a bright SN Ia (M56Ni ~ 1 M

⊙).


Patnaude+ 12

type ia supernovaeicc.ub.edu/webs/iccub_winter_meeting/sites/icc.ub... · type ia supernovae: the...

Documents