Transcript The Story of Pulsational Pair-Instability SNe

The Story of Pulsational
Pair-Instability SNe
Briana Ingermann & Parke Loyd
Meet Pulsational Pair-Instability SN
Progenitor
After a short MS life, our progenitor senses
something is wrong
A runaway thermonuclear explosion ejects
a shell, and our progenitor feels better
Fig S5 from Woosley et al. 2007 plotting the light curve resulting from the first pulse of the
110 Mʘ model..
...but the respite from instability is brief
With a final gasp, our progenitor dazzles
spectators with a display like few others
Fig 1 of Woosley et al. 2007 plotting the velocity and interior mass as a function of radius when
the second ejected shell is beginning to impact the first in the 110 Mʘ model.
Fig 2 from Woosley et al. 2007 plotting the cumulative lightcurve predicted by their simulation of a
110 Mʘ low-metallicity star.
So it goes with the pulsational pairinstability supernova
Fig 3 from Woosley et al. 2007 plotting the R-band absolute magnitude of SN 2006gy as the red data
points, predicted values with 30-day smoothing for the 110 Mʘ model as the purple curve, and predicted
values for the same model but with 4x the KE of the ejecta as the dashed curve.
Quit faking your lightcurves
When they attempt to model the
observations of SN 2006gy
they find much better
agreement when they
artificially increase the
explosion energy by a factor
of 4 (which gives an energy
above all the values in their
supplementary table 1). Is
this higher energy still
consistent with the
pulsational pair instability
mechanism they describe? If
so, what parameters of the
star would have to be
changed?
Excerpt from Table S1 of Woosley et al.
2007 showing how much the kinetic
energy and mass released in a pulse can
vary just as a function of the core mass.
Energy of the second pulse
Why is the second pulse so much higher energy than the first? Is the runaway
reaction of Helium just more efficient? Or when they say "more energetic"
do they mean energy per unit mass ejected?
•
•
The energy release they refer to in their
paper is the kinetic energy of the
ejected material. The total energy
generated by the runaway
thermonuclear reactions depends on
the specifics of those reactions (but
note a very strong dependence on
temperature).
The paper is not consistent on whether
the second pulse is always strong and
ejects less mass (see right).
But wait, doesn't the first ejected shell lose
most of its KE leaving the star while the
second hardly loses any? Why?
Excerpt from Table S1 of Woosley et al.
2007 contradicting their statement in the
caption that "in each case, the first pulse is
the weakest," as well as their statement in
the main text of the article that "later
ejections have lower mass."
Shall we compare the reactions
occurring in the first and second
pulse?
Log mass fraction
First Pulse
Interior mass / solar masses
Log mass fraction
First Pulse
Interior mass / solar masses
Log mass fraction
Second Pulse
Interior mass / solar masses
Log mass fraction
Second Pulse
Interior mass / solar masses
Comparative supernovatology
How does the driving mechanism for a PPI SN differ from a
core collapse (type II and Ib-c) or a type Ia?
Have we ever seen a SN of a
>120 Mʘ Progenitor?
Maybe:
•
•
Next week's paper (Smith et al. 2007)
suggests an initial mass of the
2006gy progenitor of ~150 Mʘ.
Cooke et al. 2012 claim to have
observed superluminous supernovae
(1051 erg) with estimated progenitor
masses of 100-250 Mʘ at high
redshift.
Plus some possible future SNe:
•
Crowther et al. 2010 claim there are
four Wolf-Rayet stars in the R136
cluster with masses of 165-300 Mʘ.
Fig. 1 from Crowther et al. 2010
showing a 12'' x 12", infrared (1.99
- 2.31 μm bandpass) image of the
R136 cluster.
Eta Carinae
Enter: the IMF
What rate would we
expect for such
events based on the
IMF? What do we
observe? I would
expect them to be
rare compared to
regular supernovae
because they require
such massive
progenitors.
Fig 2 from Cooke et al. 2012 plotting the lightcurve of a
superluminous SN at redshift z = 3.9.
How does metallicity fit in?
Our questions:
Pair instability can only occur in enormously massive stars of low to moderate
metallicity. Why this metallicity requirement?
What are the implications for observation?
And a related question from you:
They say in the paper that they use a 110 solar mass star with solar
composition, but with a mass loss rate at a fraction of the standard value.
Do they have any justification for the validity of doing this?
Why PPI?
Do pair instability ejections have to be the mechanism by which the envelope is
removed? It seems like any stellar wind followed by any explosion would
lead to high speed collisions of gas. What is special about the pair instability
ejections that make them more likely than other scenarios?
Last week we talked about WR stars, which I think are less massive than the
stars discussed, collapsing directly to a black hole with no explosion. So I
guess I am just wondering what is fundamentally different in this case which
cause these stars to make super luminous supernovae as opposed to
simply collapsing to a black hole with no explosion?
In Table 1. in the birth mass range of 130 - 260 Mʘ row they state that there is
an explosion but with no remnant. How is this possible, and if so what
happens to all the remaining mass that would have been a black hole or
neutron star? I would think that all that excess mass would have to be
directly converted to energy. This would leave at least an excess energy of
E ~ mc2 ~ (Minimum neutron star/black hole mass)•(c2) that could go into
Does the first pulse resemble a
type-II SN?
So it seems from plot S5 and figure 2 that the light curve from the first mass
ejection has a peak luminosity around 1042 ergs, which I think is around the
peak luminosity of a regular core collapse supernova. So my question is
can we detect the first outburst and what supernova types might it look
like?
When a pair-instability star first explodes and sheds its envelope, would that
explosion likely resemble a nova or a supernova itself?
Suppose that we were only able to observe the spectra of a pair-instability
supernova during one of its pulses and not its light curve. What would the
spectra of this type of supernova look like? Are there any unique features
in its spectra, apart from its extreme luminosity, that would differentiate it
from the more common types of supernovae?
Fig 2 from Kasen & Woosley 2009 plotting bolometric light curves predicted by a SN model with
varying amounts of ejected 56Ni.
Observe the SN in it's natural
habitat
Pre-SN
The paper describes how the temperature in the star after that first pairinstability explosion will determine how long it will take the star to re-ignite
and continue burning until the next pair-instability explosion (from anywhere
between days to decades [or even centuries?!]). If a star's burning
continued after a long period of time (like closer to the decades range) and
thus it didn't explode again until much later than the first expulsion of the
envelope, would we still observe the same high-energy SN like what we
saw in SN2006gy? Or could we possibly miss it, maybe suggesting that
more pair-instability SN occur but we do not detect them?
Will pulsational PISN ever go through a LBV or WR phase?
Post-SN
A PI SN site will be heavily obscured by several solar masses of ejected
material, so is it feasible to search for progenitors or remnants to these SN?
References
Cooke, J., Sullivan, M., Gal-Yam, A., et al. 2012, Nature, 491, 228
Crowther, P. A., Schnurr, O., Hirschi, R., et al. 2010, MNRAS, 408, 731
Kasen, D., & Woosley, S. E. 2009, ApJ, 703, 2205
Smith, N., Li, W., Foley, R. J., et al. 2007, ApJ, 666, 1116
Nakamura, F. & Umemura, M. 2001, ApJ, 548, 19
Table 1 from Woosley et al. 2007 showing the notional fates of stars (but at what
metallicity?).