The merging of white dwarfs?

Straw_Cat

After reading this article, I'm wondering... what happens when two white dwarfs (dwarves?) merge?
Is there sufficient mass left in both of them to create something like a new yellow star like our Sun?

http://phys.org/news/2012-08-space-w...itational.html

twofish-quant

What's believed to happen is that they produce a big explosion which is seen as a type Ia supernova.

Straw_Cat

Thanks, I appreciate the lead.
I looked up this wiki:
http://en.wikipedia.org/wiki/Type_Ia_supernova

I notice the wiki now needs an update.

Jakeus314

Type 1a is the result of a specific way a white dwarf reacts to gaining mass from another. Just don't assume that all white dwarf - white dwarf collisions result in 1a explosions.

Chronos

I was under the impression SN1a's are generally believed to be due to accretion of sufficient mass to push a white dwarf over the Chandrasekhar limit. White dwarf mergers does not appear to be a viable explanation for consistent intrinsic luminosity of SN1a's. White dwarf mergers do, however, appear to be a viable explanation for GRB's.

twofish-quant

Usually. Ia is an observational classification of a star with no hydrogen with silicon. The idea is that a small number of Ia's come through white dwarf mergers.

I don't think they are. The problem is that I don't think you can get the level of beaming that you see in GRB. The preferred model for GRB's are neutron star mergers.

Straw_Cat

The wiki states that there might be one supernova a century of paired white dwarfs. Pretty rare, if that's a correct estimate.

The discovery about white dwarf binaries discussed in the Phys.Org article might lead to a lot more attention being paid to finding these and studying them, so theories about their mergers will probably change over time.

phyzguy

There are currently two competing models for Type 1A supernovae. We're sure that the explosion is triggered by a Carbon-Oxygen white dwarf in a double star system, but it is unclear what the companion star is. In the "singly degenerate" (SD) model, the companion star is a main sequence star or red giant, while in the "doubly degenerate"(DD) model the companion star is another white dwarf. The SD model was favored several years ago, but evidence is growing that all or most Type 1As come from the DD model. This paper gives a detailed study showing that the DD model can certainly explain Type 1As, while the Nature paper I've attached on the recent Type 1A SN2011fe, shows that no reasonable MS or RG star could be found before the explosion, meaning that this event must have come from a DD merger.

Attached Files

SN1A_nature10646.pdf (632.4 KB, 5 views)

phyzguy

I think one Type 1A per century per galaxy is about the right rate to explain the Type1A's that we see.

There are already a large number of people studying this question. I think you may have this backwards - these white dwarf binaries were probably discovered in an effort to quantify the rate of white dwarf binary mergers to see if they explain Type 1A's.

Chronos

The super Chandresekhar mass SN1a's that occur in the double degenerate model appear to muddy things up. I have to think this would lead to significantly inconsistent brightness between various SN1a.

phyzguy

People refer to Type-1A supernovae as "standard candles", but really what they mean is that they are "calibratable standard candles". In practice Type-1A peak absolute magnitudes vary by about 3 magnitudes, or more than an order of magnitude in luminosity. People use what is known as the "Philips curve" to calibrate the absolute magnitude based on the rate at which the magnitude drops off with time. Modeling, like the paper I linked earlier, seems to suggest that white dwarf mergers can be consistent with this range in brightness. The jury is still out, but the case that most Type-1A's are WD mergers seems to be getting stronger all the time.

twofish-quant

And yes, the possibility that this throws off the calculations of the accelerating universe has occurred to people. What people have done to have some confidence that this isn't the situation is to try to correlate brightness with other supernova Ia characteristics. The big one that people are worried about is composition. Main sequence stars that are closer have more metals than those that are far away. It is possible that this affects brightness? Yes, it's possible and the original SN cosmology papers put quite a bit of effort into trying to rule this out.

For example, you divide your supernova into two groups based on some spectral characteristic. If there was some systematic difference in supernova brightness, then the curves would be different, but they aren't. Once you've convinced yourself that there isn't any systematic difference in supernova Ia, then what is left is "random error" and the random error in supernova Ia brightness is than the signal of cosmic acceleration.

It's not totally airtight and while unknowns in supernova Ia evolution doesn't wipe out the signal for cosmic acceleration it does add noise which limits your precision, which is why people are interested in really 1) understanding what is going on with supernova Ia and 2) coming up with cosmological measures that have nothing to do with supernova.

twofish-quant

The other thing is that it's not hard to get consistent white dwarf luminosities in a DD scenario. You can wave your hands and argue that only a small set of stellar situations will result in rotating white dwarfs.

The other thing is that the delayed-detonation mechanism (the last time I checked which was several years ago) turns out to be quite robust and insensitive to parameters.

One cute thing is that I knew of an SNIa researcher that did some work for a car company. The situation in which rapid burning creates a shock wave is known as "pinging" and it's something car makers try to avoid.

twofish-quant

There may be a pendulum, because when I was in grad school (in the 1990's), it seemed that the issue was settled in favor of DD, and I was a little surprised when I did some research in this thread and found out that people were still talking about SD.

Also, this might make a good mini-research project for some undergraduate. If the OP is game, it would be nice to spend a few days looking at papers and summarize the current state of the topic and how the consensus changed over time (if it has).

Chronos

And there is yet another contender - the core degenerate [CD] model. The big picture remains remains muddy, IMO. All these models make sense and probably are correct at some level. I like TRGB distance calibration, it makes sense to me. I am a little worried about the low value for Ho. It suggests we may be missing something important.

SHISHKABOB

I believe it might be Neutron star mergers that are one of the viable explanations for (I think) short duration GRB. Of course, I am no expert on the subject, it is just through my readings on GRBs that I heard about Neutron star mergers causing them.

twofish-quant

This points out one of the "interesting gaps" in our understanding of stellar evolution. If you look at an intro astronomy textbook they have the standard story for how a star evolves. However, that story is for single stars, and most stars are in multiple-star systems.

We know from cluster observations that multiple stars evolution is close to single star evolution, but SN Ia and GRB's are examples where multiple star evolution might be very different.