Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I What % of the heavy elements are produced by kilonovas vs. supernovas?

  1. Nov 1, 2017 #1
    So the recent neutron star merger event showed that most of the heavy elements (gold, platinum, uranium, etc.) are produced in kilonovas. But with neutron star mergers so rare, there can't be that many kilonovas. Prior to this I always used to think they were mostly produced in supernovas. The chances that a solar system was seeded by a supernova seem much higher than one seeded by a kilonova. What proportion of heavy elements are produced in kilonovas vs. supernovas?
     
  2. jcsd
  3. Nov 2, 2017 #2

    stefan r

    User Avatar
    Gold Member

    The mergers do not need to be that common. Compare magnesium at 6 x 10-4 abundance in the universe to osmium at 3 x 10-9. Some of the material ejected by core collapse supernovas is still hydrogen and helium.

    The Milkyway's core has several million solar masses. I do not know how much of that came from neutron stars.
     
  4. Nov 4, 2017 #3
    I am at a loss to understand how two objects made of degenerate matter can produce any element by merging. When all the protons and electrons have been combined to make neutrons, how can any other element be made after that?
     
  5. Nov 4, 2017 #4
    Because when you tear apart the neutron star, neutrons would start beta decay.
     
  6. Nov 4, 2017 #5
    Okay, but a merger of two neutron stars does not tear apart either of the neutron stars. Some material may be expelled - and that is certainly true in the case of the GRBs, but the majority of the degenerate matter is going to be combined to either form a larger neutron star or possibly a black hole if there is sufficient mass. It is nothing like supernovae, which do blow themselves apart.
     
    Last edited: Nov 4, 2017
  7. Nov 4, 2017 #6
    Far from all types of supernovas result in the star blowing itself up completely.
     
  8. Nov 4, 2017 #7
    That's an interesting point.
    Even if there were in mergers, unknown conditions which lead to faster neutron decay, all that is going to do is generate hydrogen (I suppose)
     
  9. Nov 4, 2017 #8

    phyzguy

    User Avatar
    Science Advisor

    You're correct in that the majority of the degenerate matter remains part of the larger neutron star or collapses into a black hole. The estimates of the recently observed NS-NS merger were that only between 0.1% and 1% of the original neutron star matter was expelled. However, that tiny fraction is enough to explain the amount of heavy elements here on Earth. Elements like gold are billions of times less abundant than hydrogen.
     
  10. Nov 4, 2017 #9

    phinds

    User Avatar
    Gold Member

    Good point but doesn't address the core of the OP's question.
     
  11. Nov 4, 2017 #10

    phyzguy

    User Avatar
    Science Advisor

  12. Nov 4, 2017 #11
    ... but neutron decay does not have to mean Hydrogen?
    It could produce any element if unconfined quarks are allowed?
     
  13. Nov 4, 2017 #12

    phyzguy

    User Avatar
    Science Advisor

    Is this in response to my Post #8? I didn't mean to imply that the decompressed neutron star matter gives rise to hydrogen. It gives rise to heavy elements. My point was that heavy elements like gold are very rare, so if only 0.1% of the neutron star matter is ejected and converted to heavy (r-process) elements, that is enough to explain the observed abundance of these elements.
     
  14. Nov 4, 2017 #13
    Thanks for your link about the r process.
     
  15. Nov 5, 2017 #14
    It is extremely difficult to believe that all the gold, platinum, uranium, etc. in the universe was created only by neutron star mergers. Considering that neutron stars themselves are already rare, and mergers of neutron stars are exceedingly rare, and you are talking about only between 0.1% and 1% of that ejected degenerate material creating everything heavier than iron in the universe. I realize that elements heavier than iron are not common, but even a billion times rarer than hydrogen is still more common than the process I just described. That process could not possibly account for all the elements heavier than iron in the universe.

    FYI, according to the paper below the NS-NS merger ejected between 0.03 to 0.05 solar masses of material, which puts the ejecta slightly higher than 1% of the combined mass.

    Spectroscopic identification of r-process nucleosynthesis in a double neutron-star merger - Nature 551, November 2017
     
    Last edited: Nov 5, 2017
  16. Nov 5, 2017 #15
    When the neutron stars merge, the looser upper crusts of these objects are not as tightly coupled with the rest of the object and they have a potential of getting loose into space. These loose pieces or shards of neutrons are no longer gravitationally bound tightly enough to remain as degenerate matter, so they'll (explosively) break up into regular atomic nuclei. Then some of the neutrons will beta decay into protons and electrons along the way too, so they won't stay neutrons forever.

    Each of the neutron shards might be of various sizes, with many different numbers of nucleons per shard. They beta decay, and fission down to the most stable configurations in the end.
     
  17. Nov 5, 2017 #16
    So one of the articles I've recently read is this one:
    Neutron star collisions may have created most of the gold in the universe | Popular Science

    There is a graphic in there, which shows which percentage of elements come from what types of explosions:

    periodic_table.jpg

    There are 6 types of origins mentioned: (1) Big Bang, (2) cosmic ray fission, (3) merging neutron stars, (4) exploding massive stars, (5) dying low-mass stars, and (6) exploding white dwarfs. Now #1 through #4 are pretty clear what they are, but #5 & #6 seems a little unclear, wouldn't these be the same things? The whole graphic is a bit confusing. It says high mass star explosions produce only the elements upto Zr, Zirconium. All heavier elements are produced by either dying low-mass stars, or neutron star mergers! This makes no sense whatsoever!
     
  18. Nov 5, 2017 #17
    Thanks for the explanation. snorkack also explained that process to me. It would certainly appear to be one method for creating elements heavier than iron, but considering the rarity of the event and the small amount of mass they eject during that merger there has to be another process to explain the elements heavier than iron.

    Even supernovae are rare, on average only one supernova per galaxy per century, and yet every neutron star is the result of a supernova. Supernovae eject vast quantities of material (particularly Type II supernovae), many times the mass of our sun and they do so at relativistic speeds. Why does it have to be one or the other, could not both processes contribute to the elements heavier than iron in the universe?
     
  19. Nov 5, 2017 #18
    Neutron star mergers might have been more common in the early universe, maybe? Back in the first generation of stars, pretty much all of them were massive enough to produce neutron stars or black holes after they went supernova. Then I'm guessing that these early galaxies were filled with heavy gas clouds (unlike today's galaxies) that might have provided a braking force against speeding neutron stars, and then they might have gotten caught up within gas clouds and several neutron stars might have found themselves in gravitational locked to each other, and they quickly merged (took millions of years rather than billions). Seeding the early universe quickly.
     
  20. Nov 5, 2017 #19
    By "dying low-mass star" I presume they mean a star smaller than ~3 solar masses, but larger than ~0.75 solar masses, such as our sun. Stars smaller than ~0.75 solar masses haven't come close to dying yet. An "exploding white dwarf" would refer to either a Type Ia or Iax SNe. In the case of the Type Ia SNe the white dwarf is completely obliterated. In the case of Type Iax SNe some of the white dwarf remains.

    It was also my understanding that when "low-mass stars" die they begin to lose their gravitational hold on their outer layers and the material is ejected to form planetary nebula. However, that ejection of material is not at relativistic speeds.
     
  21. Nov 5, 2017 #20
    No. It would generate helium 3.
    A neutron undergoes beta decay - half-life 10 minutes.
    The resulting proton captures a neutron - D is stable.
    D captures a neutron... and T cannot capture another neutron. Because H-4 would be an unbound state - any extra neutrons just bounce off the tritons until they undergo free neutron beta decay and the one proton formed binds another 2 neutrons.
    T is radioactive... with 12 years halflife, eventually forming He-3.
    He-4 is also at dripline - neutrons bounce off with no effect because the third neutron of He-5 is unbound.
    But now suppose that somehow there are 3 protons together. Say stable Li-7.
    What next?
    Li-8 and Li-9 are both bound. Only Li-10 is not.
    The half-lives are Li-8 840 ms, Li-9 178 ms. Drastically faster that the 600 s half-life of neutrons
    And Li-9 forms (50 % of times) stable Be-9.
    Repeat with Be. First unbound species Be-13. So neutron dripline at Be-12. Halflife 21 ms.
    Etc. All the way to over 100 protons.
     
  22. Nov 5, 2017 #21
    Without a doubt neutron star mergers were much more common in the early universe than they are today. Population III stars were much larger than Pop. II and I stars, so the percentage of neutron stars would have been proportionally increased - as would the number of supernovae in the early universe. Both combined could explain the elements heavier than iron, but neutron mergers on their own wouldn't cut it. I also don't understand how "low-mass stars" can contribute any element heavier than iron since they do not eject their outer envelope at relativistic speeds and it is not degenerate material being ejected.
     
  23. Nov 5, 2017 #22

    phyzguy

    User Avatar
    Science Advisor

    So you have done the calculation of the amount of r-process elements which are created in these events and compared it to the observed abundances? If so, please share that calculation with the rest of us. If not, maybe you shouldn't state the conclusion until you have done the calculation. The people at LIGO in this paper have done the calculation, and they conclude (see a quote from that paper below) that these events can produce all of the observed r-process elements. Since you have a different conclusion, can you please explain what they have done wrong?


    "Our results suggest that dynamical ejecta from rare NS mergers could be an important and inhomogeneous source of r-process elements in the galaxy (Ji et al. 2016 ;
    Beniamini et al. 2016 ). If more than 10% of themass ejected from mergers is converted to r-process elements, our prediction for average r-process density in the local universe is consistent with the Galactic abundance."
     
  24. Nov 5, 2017 #23

    phyzguy

    User Avatar
    Science Advisor

    Perhaps another error in your thinking is that neutron stars are not as rare as you seem to think. NASA estimates that there are on the order of 1 billion neutron stars in our galaxy.
     
  25. Nov 5, 2017 #24
    Well, then NASA has a problem. Because they also estimate that there are on average of three supernovae every century in the Milky Way galaxy. Even if every supernovae in the Milky Way from the beginning of time until today became a neutron star it would not be nearly enough to be one billion neutron stars. In order for there to be a billion neutron stars in the Milky Way galaxy there would have to be at least one supernova in the Milky Way galaxy on average every 13.8 years, and every supernova would have to result in a neutron star. Clearly one, or both, of those estimates is flat out wrong.
     
    Last edited: Nov 5, 2017
  26. Nov 5, 2017 #25

    phyzguy

    User Avatar
    Science Advisor

    If the current supernova rate of about 1 every 33 years had existed for all of cosmic history, and the Milky Way is about 10 billion years old, then we would expect there to be about 10 billion/33 = 333 million neutron stars. This is not so different from the 1 billion order of magnitude estimate I gave. However, it is well known that this is not the case. Looking at the image below, from this site, we can see that in the past stars were forming (and dying!) at a rate more than 10X what we see today. Thus there are many more "dead" stars, like neutron stars than you would predict by your simple calculation. I urge you to learn some astronomy before you start attacking what is known.


    vs_sfr.png
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted