High-resolution simulation of a galaxy hosting a super-luminous supernova and its chaotic environment in the early Universe. Credit: Adrian Malec and Marie Martig (Swinburne University)
Some of the earliest stars were massive and short-lived, destined to end their lives in huge explosions. Astronomers have detected some of the earliest and most distant of these exploding stars, called ‘super-luminous’ supernovae — stellar explosions 10–100 times brighter than other supernova types. The duo sets a record for the most distant supernova yet detected, and offers clues about the very early Universe.
“The light of these supernovae contains detailed information about the infancy of the Universe, at a time when some of the first stars are still condensing out of the hydrogen and helium formed by the Big Bang,” said Dr. Jeffrey Cooke, an astrophysicist from Swinburne University of Technology in Australia, whose team made the discovery.
The team used a combination of data from the Canada-France-Hawaii Telescope and the Keck 1 Telescope, both located in Hawaii.
“The type of supernovae we’ve found are extremely rare,” Cooke said. “In fact, only one has been discovered prior to our work. This particular type of supernova results from the death of a very massive star (about 100 – 250 times the mass of our Sun) and explodes in a completely different way compared to other supernovae. Discovering and studying these events provides us with observational examples to better understand them and the chemicals they eject into the Universe when they die.”
Super-luminous supernovae were discovered only a few years ago, and are rare in the nearby Universe. Their origins are not well understood, but a small subset of them are thought to occur when extremely massive stars, 150 to 250 times more massive than our Sun, undergo a nuclear explosion triggered by the conversion of photons into electron-positron pairs. This process is completely different compared to all other types of supernovae. Such events are expected to have occurred more frequently in the early Universe, when massive stars were more common.
This, and the extreme brightness of these events, encouraged Cooke and colleagues to search for super-luminous supernovae at redshifts, z, greater than 2, when the Universe was less than one-quarter of its present age.
“We used LRIS (Low Resolution Imaging Spectrometer) on Keck I to get the deep spectroscopy to confirm the host redshifts and to search for late-time emission from the supernovae,” Cooke said. “The initial detections were found in the CFHT Legacy Survey Deep fields. The light from the supernovae arrived here on Earth 4 to 6 years ago. To confirm their distances, we need to get a spectrum of their host galaxies which are very faint because of their extreme distance. The large aperture of Keck and the high sensitivity of LRIS made this possible. In addition, some supernovae have bright enough emission features that persist for years after they explode. The deep Keck spectroscopy is able to detect these lines as a further means of confirmation and study.”
Cooke and co-workers searched through a large volume of the Universe at z greater than or equal to 2, and found two super-luminous supernovae, at redshifts of 2.05 and 3.90 — breaking the previous supernova redshift record of 2.36, and implying a production rate of super-luminous supernovae at these redshifts at least 10 times higher than in the nearby Universe. Although the spectra of these two objects make it unlikely that their progenitors were among the first generation of stars, the present results suggest that detection of those stars may not be far from our grasp.
Detecting the first stars allows us much greater understanding of the first stars in the Universe, Cooke said.
“Shortly after the Big Bang, there was only hydrogen and helium in the Universe,” he said. “All the other elements that we see around us today, such as carbon, oxygen, iron, and silicon, were manufactured in the cores of stars or during supernova explosions. The first stars to form after the Big Bang laid the framework for the long process of enriching the Universe that eventually produced the diverse set of galaxies, stars, and planets we see around us today. Our discoveries probe an early time in the Universe that overlaps with the time we expect to see the first stars.”
Sources: Keck Observatory, Nature
The comoving distance out to the z = 3.9 SN is 22 billion light years. This occurred 1.5 billion years into the evolution of the universe, or 12.2 billion years ago.
This is not likely a typeIII star, unless a few stars composed almost entirely of hydrogen and helium still formed then, making latent typeIII stars.
LC
Wouldn’t these two super-luminous supernovae be considered hypernovae?
http://en.wikipedia.org/wiki/Hypernova
It is interesting that my conjecture about PopIII stars appears to have not been too far off what has been considered by astrophysicists.
LC
The description of SN 2006gy says that the entire star is destroyed by the pair instability explosion, so where did the black holes come from in the early universe?
Ah, I see that other supermassive stars undergo photodisintegration which is endothermic for anything less than iron so they collapse into black holes. If the formation of POPIII stars is consistent then they would have matured in a very small time frame and give us an early universe with both sets of stars spraying metals everywhere and one set creating black holes to pull the stuff in to the Eddington limit, which could explain the mature galaxies with high redshifts. Once we can see back far enough will it look like 4th July/5th November?
Back in the 1960s there was work on something called the hyperstar. This was a star with an unusually large mass, up to a billion solar masses. For various reasons the idea did not work out. However, varieties of PopIII stars could be a small version of a hyperstar. This is a H and He gas cloud that accretes at the center and generates fusion. The H and He have a low opacity so there is a weak thermal force which opposes the self-gravitation of this object. I think not a lot is understood about these, where it is conceivable they reach thousands of solar masses.
The core of this small hyperstar could be a complicated system that ends up producing a black hole surrounded by a shell of H and He that fuses in a sustained or maybe an intermittent fashion. This could exhibit a sequence of very violent events. The main different from a standard accretion disk around a black hole is that hydrogen and helium weakly scatter or are thermalized by huge outbursts of radiation. So this material can keep accreting inwards. The outer core would then be some shell or disk of material that has a lot of fusion occurring. The inner core might be a black hole, and if this object could accrete tens or hundreds of thousands of solar mass units of hydrogen and helium, then even if a small percentage of it produces a black hole it could result in a 100 to 1000 solar mass black hole. It might also breed up the earliest heavier elements and form sufficient amounts for the formation of more normal PopII stars.
LC
Thanks, that’s a really intriguing idea, a giant “star” shell around a black hole with a combination of accretion and unstable shell fusion. Eddington limit versus gravity and thermonuclear or even degenerate pressure I wonder if those black holes have evaporated yet and if not, would they have a special signature…
The black hole evaporation time is t ~ M^3, and is about 10^{67} years for a solar mass black hole. So the black holes considered here would not be near evaporation.
LC
Also, any formula of evaporation would only be acurate for a non-feeding blackhole. They can grow also, and combine, and who can say not explode.
The formula for the duration of a black hole
t = 5120?G^2M^3/?c^4 = 8.4×10^{-17}M^3 kg^{-3}sec
where ? is the Planck unit of action of quantum physics. For the sun with M = 2×10^{30}kg the duration of a solar mass black hole is 6,7×10^{74}sec or 2.1×10^{67} years. This assumes of course a black hole in a vacuum.
One can compute the more general problem. If we assume a black hole exists in a spacetime with a temperature T and the black hole has a mass M so that the same temperature T = ?c^3/8?GMk, k = Boltzmann constant. If the black hole absorbs a photon from the background it mass increases M — > M + ?M and its temperature correspondingly decreases. This increases the probability it will absorb another photon rather than emit one and the black hole statistically is likely to grow. If the black hole emits a photon from the background it mass decreases M — > M – ?M and its temperature correspondingly increases. This increases the probability it will emit another photon rather than absorb one and the black hole statistically is likely to shrink. The funny conclusion is that equal temperature does not mean equilibrium. BTW, the temperature of a solar mass black hole would be around 6,8×10^{-8)K. A black hole with about the mass of the moon would have the same temperature as the CMB temperature of the current state of the universe.
One can compute a situation where there is a small black hole that emits and absorbs particles in a way that is sustained. One could imagine a sort of mini-black hole reactor that feeds matter into a tiny black hole and uses the radiation emitted as an energy source. This would be a tricky bit of technology to control, and it would be dangerous.
LC
The ‘lower’ mass pair-instability SNe (150-250Msun, highly uncertain) are expected to leave no core remnant, complete unbinding.
If stars can become even heavier, maybe above 300Msun, they would seem to collapse to a quite large black hole (all of 300+ M) without any explosion at all, a complete collapse.