Superbright Supernova First Observed of Antimatter Variety

Article written: 2 Dec , 2009
Updated: 24 Dec , 2015


The supernova 2007bi wasn’t your typical supernova: it was 10 times brighter than a Type Ia supernova, making it one of the most energetic supernova events ever recorded. Astronomers from the University of California Berkeley have analyzed the explosion, which was recorded by a robotic survey in 2007, and found that it is likely the first confirmed observation ever made of a pair-instability supernova, a type of extremely energetic supernova that has been theorized but never directly confirmed.

The confirmed observation of a pair-instability supernova has been long-awaited – the theory that they exist has been around since the 1960’s – but it appears as if the wait is over. The supernova 2007bi, seen by the Nearby Supernova Factory in April of 2007, is the first observed supernova that fits the bill for the unfathomably huge proportions of pair-instability supernovae explosions. A team of astronomers led by Alex Filippenko of the University of California Berkeley published their analysis in in the December 3rd issue of Nature. The discovery was initially made by the Nearby Supernova Factory, and emission spectra of the event was taken with the Keck Telescope and Very Large Telescope in Chile 

These type of supernovae occur only in stars above 100 solar masses, and are incredibly bright. Energetic gamma rays are created by the intense heat in the core of the star. These gamma rays, in turn, create antimatter pairs of electrons and positrons. Because of this antimatter production, the outward pressure exerted by the nuclear reactions in the core of the star is lessened, and gravity takes over, quickly collapsing the massive core of the star and creating a supernova.

There are theorized to be two kinds: those that explode with just enough force to allow for the mass around the leftover core of the star to recombine, and those that explode completely with not a smidgen left to form a black hole or neutron star. The supernova 2006gy, which had a luminosity 10 times that of a Type Ia supernova, is thought to be of the first variety. Here’s our story on that one, Could Antimatter be Powering Super-Luminous Supernovae? and Eta Carinae may also fit the profile.hese types of pair-instability supernovae will eject the outer shells of the star’s matter, settle down into an equilibrium, and repeat that process until the mass is low enough for a normal supernova to occur.

But 2007bi was much too massive to settle back down and explode multiple times. With a mass of 200 suns, the runaway thermonuclear explosion that happened in its core was energetic enough to effectively vaporize the entire star. Pair-instability supernovae in stars above 130 solar masses leave nothing behind in the way of black holes or neutron stars, but because they are so energetic and luminous, the increasing light from the explosion peaks over a very long time – 70 days in the case of 2007bi.

Though the team detected the supernova almost a week after the peak, they were able to calculate the duration of the light curve. They then studied the remnants of the explosion over the next 555 days as it faded away.

Filippenko said, “The central part of the huge star had fused to oxygen near the end of its life, and was very hot. Then the most energetic photons of light turned into electron-positron pairs, robbing the core of pressure and causing it to collapse. This led to a nuclear runaway explosion that created a large amount of radioactive nickel, whose decay energized the ejected gas and kept the supernova visible for a long time.”

The star was unique in another way: it lies in a nearby dwarf galaxy, which contains little else but the elements hydrogen and helium. Because of this, 2007bi is much like the stars that existed near the beginning of the Universe, before the trillions of supernovae populated the Universe with heavier elements. Looking more closely at dwarf galaxies – the Universe has them in spades, but they are quite dim – may be the key to observing more supernovae of this kind. Being able to study its explosion and aftereffects will give scientists a look into what the earliest massive stars acted like.

Source: Berkeley Lab press release


19 Responses

  1. tacitus says

    Back in 2007 Alex Filippenko (the lead astronomer on this supernova paper) videotaped all 42 of his lectures for UC Berkley’s “Introduction to Astronomy” class. If you haven’t seen him speak, he is a first class communicator and his love for astronomy (and live in general) comes across to the students clear and loud in his lectures.

    I would recommend watching the entire series to anyone who would like to know more about the subject. It is a fantastic and entertaining introduction to astronomy, and will also be enjoyed by people who already follow the latest on space and astronomy.

    You can find the lectures, free of charge, on the following UC Berkley web site:

    There is some routine class business at the start of the first lectue (and some of the others) but you can easily skip past that to the good stuff.

  2. Hon. Salacious B. Crumb says

    Are these not formally called Hypernovae?

  3. Hon. Salacious B. Crumb says

    Just a point. The Berkley Press release misses that point somewhat. Such “hypernova” by definition are one hundred times more luminous than ordinary supernova.

    It also mention in the graphic of the prodigious emission of gamma rays (and unobservable positrons, but their annihilations in the visible super-illumination of expanding shell).

    What worries me is the lack of gamma-ray observation from the event, and their seems no correlation between any gamma ray event to the visual observations.

    The other issue is that 200 solar mass stars almost certainly form black holes, which should be produced in its heart. They don’t discuss the alternative possibility of black hole formation and showing an expanding shell.

    There is more here than the Berkeley story than is being said.

    A great and important story, Nicholos

  4. Hon. Salacious B. Crumb says

    My point is that the gamma ray observations would be conclusive proof of the hypernova scenario.

    As yet, after reading the story, I have not found anything about it.

    Until we know if this is true (or not), the whole story is incomplete. (It might be why they still refer to the event as a “supernova”, here!!)

    I meant no suggestion what they say is wrong, nor the conclusion is flawed.

  5. Hon. Salacious B. Crumb says

    At the S&T site,
    they report that

    “But for one of the originators of PISN theory, the case is not yet ironclad. “I give it a 50% chance of being a pair-instability supernova,” says Stan Woosley (University of California, Santa Cruz). “The case has to be very tight. I’m not sure yet that it is, though in deference to the authors, I think they have tried and passed every test they could think of.”

    One crucial point that I’ve missed (and not stated here) is “SN 2007bi’s host galaxy is 1.6 billion light-years from Earth,”

    Gamma-ray observations might not have been observed from so far away! Pity

  6. Hon. Salacious B. Crumb says

    Just a final post (apologies if I look like I’m taking over the story)

    It is worthwhile and recent published arvix paper ;

    Maurer, J.I., Characteristic Velocities of Stripped-Envelope Core-Collapse Supernova Cores given out on 19th November 2009. (SN 2007 bi is only partially mentioned)

  7. Hon. Salacious B. Crumb says

    The definitive paper on detecting this class of object is; Scannapieco, E., Madau, P., Woosley, S., Heger, A., Ferrara, A.,”The Detectability of Pair-Production Supernovae at z <~ 6” ApJ., 633, 1031 (2005) @

  8. Manu says

    Two questions:

    – if the star is at the oxygen stage, _and_ so hot, what keeps oxygen to fuse into the heavier elements and go the usual way to iron?

    – the pairs will likely recombine and give back their energy as gamma rays: shouldn’t this stop the collapse?

    Where did I get it wrong? 😉

  9. DrFlimmer says

    @ Manu:

    The oxygen does not have the time to fuse further. The star destroys itself first.
    On the other hand, of course you will have a bit of iron already. But a big portion of oxygen will remain as is.

    Definitely will the positron recombine with some other electron. But what’s critical here, is time. The gamma rays will all, more or less spontaneously, become e-p-pairs at the time (in our scales it should really be no more than a blink of an eye). Thus, the core of the star loses its stability instantaneously and collapses. The recombination starts a moment later – but this moment is a moment too long.
    This is, what I would think about it. But I’m not an expert here. Maybe some others (like HSBC?) can give some better answers.

    @ HSBC:

    Very interesting posts! I will take a look at the links when I’m back home, tonight.

  10. Lawrence B. Crowell says

    Wow, this is pretty incredible.

    The researchers did not detect gamma ray production because that occurs deep in the interior. The physics from a astrophys-101 perspective is this. Particles with mass have what we call in quantum field theory a mass-gap in their quantum field physics. It requires an energy E = E’ + 2mc^2 to generate pairs of particles, say electron-positron pairs, of mass m, and where the E’ is an additional energy required to get threshold of production. This process will not increase the energy or temperature of the core, but reduce it. The reason for this is something called the equipartition theorem in statistical mechanics. The energy of a system in a thermal distribution is E = NkT. The N here is the number of particles and k is the Boltzmann constant. Now in the production of particles we have that there are now (2 + N)kT, and if a lot of these particles, say n of them, are generated then E = NkT goes to E = (N + n)kT. This does not produce energy, for energy in the motion of particles in the hot gas is being used to generate these particle anti-particle pairs. So E remains constant here, at least in the adiabatic assumption I am employing. Well this means

    NkT = (N + n)kT’

    for T’ the new temperature which results from the pair production. It is then clear that T’ = TN/(N + n) < T. So the temperature decreases!

    Well what happens when the temperature of the stellar core decreases? We appeal to the natural gas law pV = NkT, where if the temperature decreases then pV increases. The pressure is what sustains the material against the gravitational force from imploding the star inwards. For the star with a mass M = rho*V (rho = density of material) the differential of gravitational pressure on a unit of material at radius r is dp = (G*rho/r^2)dr, which can be easily integrated for a constant rho, but rho may change with radius. The upshot is that V decreasesm which then increases the pressure, which then adjusts the temperature back up, maybe to a higher temperature. This then promotes more pair production, causing more collapsing and greater pressure, causing even higher temperatures and so forth. The whole thing runs away on itself and generates this huge explosion or hypernova.

    The author Paul Preuss of the Berkeley report writes …” recorded SN 2007bi before its peak brightness and could provide enough data to calculate the duration of the rising curve, an extraordinarily long 70 days – more evidence for the pair-instability identification.” This is the apparent signature of the physics I discuss above. Apparently when worked out in detail the runaway situation I argue for above is a lower process than a standard supernova, but with a huge net production of energy.

    Lawrence B. Crowell

  11. Hon. Salacious B. Crumb says

    “The researchers did not detect gamma ray production because that occurs deep in the interior.”

    What you say here is correct, but I thought because of the size of the star the initial collapse is transparent to gamma-ray -forming initial gamma ray burst. Star of +100 solar mass would appear as a very bright and gamma-ray transient source, then would be closed off (so to speak) when the density increases sufficiently, then starting the cascade. Also 100 to 200 solar mass stars are much bigger in area and behave differently because of their relative purity of hydrogen and helium.)

    After that. I then though the core region collapses and starts the long cascade of reactions – as stated above, and the eventual formation of nearly a solar mass worth of nickel and iron.

    Of course the exciting thing is this explains the original of the heavier elements in the earlier stages of the universe.

    Regardless, hypernovae make supernovae more like powder puffs!!

  12. Manu says

    DrF, LBC: thanks, fascinating stuff!

  13. Lawrence B. Crowell says

    Hon. Salacious B. Crumb

    The Berkeley article mentions that the earliest stars might be of this nature. PopIII stars are largely made of hydrogen and very transparent to radiation up to X-rays. So secondary Compton scattered photons from pair production might be directly observable. We will have to go from z = 6 to z > 8 to get data on these.


  14. Hon. Salacious B. Crumb says

    @ Lawrence

    You make an interesting point. The relationship of “hypernovae” to gamma ray burst has been studied mostly around 2004.

    Much of my ideas expressed here come fro the paper ; Podsiadlowski, Ph., “The Rates of Hypernova and Gamma-Ray Bursts : Implications for their Progenitors“; A.J., 607, L17, 2004

    There are several papers like this one.

  15. Hon. Salacious B. Crumb says


    Another very useful one that I’ve read is;

    “Gamma-Ray Bursts, Collapsars and Hypernovae”

  16. Lawrence B. Crowell says

    Thanks for the reference. In the first paper the author talks about GRB beams with gamma = 100 or more. That is considerable. I’ll see what I can make of this. There seems to be a lot of reference to astrophysics that I am probably not familiar with.

    I’ll try to do that before Anaconda rubbishes up this blog thread.

    Cheers LC

  17. Member
    IVAN3MAN says

    @ Hon. Salacious B. Crumb,

    RE: “Characteristic Velocities of Stripped-Envelope Core-Collapse Supernova Cores”.

    The link that you provided above for that paper does not work; it is the URL for this Universe Today post.

    The correct link is here.

    P.S. Thanks for the other links.

  18. Hon. Salacious B. Crumb says

    @ IVAN3MAN

    Thanxs for the post here. I looked at the original text i had sent and the address was sent (not universe today). Strange. I wonder if it was changed by the site here?

    anyway. Cheers for the proper link!!

  19. RUF says

    OK — I’m missing something here…

    Type IAs occur at 1.44 solar masses. How can one occur in a star 100 times bigger than the sun?

Comments are closed.