Royal Astronomical Society Press Notices
Tuesday 8th April 1997

Solved: The Mystery of the Star That Blew Up as Supernova 1987A

Ten years after Supernova 1987A in the Large Magellanic Cloud became the first naked-eye supernova in our skies since 1604, Oxford astronomer Philipp Podsiadlowski says that the mystery about the nature of the star that produced this extremely unusual supernova is probably solved. The star that exploded used to be a member of a close binary pair, but it merged with its partner in an act of stellar cannibalism in the relatively recent past -- perhaps some 30,000 years ago. Dr Podsiadlowski will explain how this scenario can account for all SN1987A's oddities when he speaks on Thursday 10th April at the UK's National Astronomy Meeting taking place at the University of Southampton.

One of the main puzzles of SN1987A, recognized soon after the explosion, is the fact that the star that exploded was a BLUE supergiant (with a radius of about 40 times the radius of the Sun), whereas theory had predicted that massive stars end their lives as RED supergiants (more than 1000 times larger than the Sun). One early idea was that this behaviour was somehow connected with the fact that the Large Magellanic Cloud, a satellite galaxy of the Milky Way, has a somewhat lower concentration of heavier chemical elements than the Milky Way, and that the star used to be a red supergiant but turned into a blue supergiant just 30,000 years ago. But the most recent calculations have shown that this idea just cannot be made to work. Even if it did work, it still would not explain the two other main puzzles of the supernova: the fact that a significant fraction of the material from the core of the star seems to have been thoroughly mixed with its outer layers, and the complex nebula surrounding the supernova remnant.

The nebula around the supernova remnant was first discovered with the European Southern Observatory's New Technology Telescope, but is most clearly seen in images taken with the Hubble Space Telescope. The main nebula consists of three rings that appear to float in space. What is left of the supernova is at the centre of the inner ring (with a radius of 0.7 light years), while the two other rings are parallel to the inner ring, but displaced from it by about 1.5 light years above and below. In the north, the nebula is bounded by a structure that has the appearance of Napoleon's hat.

The whole of this nebula was created before the supernova went off and consists of matter that was ejected by the progenitor star (or stars) within the last 30,000 to 60,000 years. It is very noticeable that the nebula is distinctly non-spherical, but nevertheless is symmetrical around an axis. This suggests that the progenitor was a spinning very fast around that axis. The most recent Hubble Space Telescope image of the supernova provides further evidence for this idea, since for the first time it resolves the actual material ejected during the supernova explosion and shows that it is elongated in the direction perpendicular to the inner ring. This is expected if the star that exploded was not spherical but flattened because it was spinning rapidly. But the rules of physics say that no single star, even if it were born with rapid rotation, could still be turning so fast after it has expanded to supergiant dimensions.

However, the conundrum can be resolved if something acted to 'spin up' the progenitor before it exploded. It turns out that a binary companion is capable of doing just that. Indeed, the type of binary system required is quite typical -- most stars are members of binary or multiple systems. The only important requirement is that the merger of the two stars takes place after the progenitor has consumed all the helium in its core.

During the merger, which itself takes only a few years or decades to complete, the companion star is completely destroyed and its material is mixed with the envelope and part of the core of the progenitor. This produces a rapidly rotating star with thoroughly mixed outer layers -- thereby explaining that observed peculiarity. This star subsequently wants to shrink to become a blue supergiant. However, the merged system is rotating too rapidly to make a blue supergiant and needs to slow down. This it achieves by spinning off matter in a disc around its equator. Once the star has become a blue supergiant, it develops a powerful stellar wind that sweeps the disc outwards. This process forms the observed inner ring. The outer rings may be 'swept-up' parts of a shock region in the shape of a double cone, produced by the interaction of stellar winds from each of the two stars before they merged.

It now seems that a double-star merger scenario is the only way in which all the various anomalies of this very unusual supernova can be understood. This theory predicts particular chemical anomalies, which would have been produced during the merger itself. If these are detected, it would be virtually conclusive evidence that the theory is correct.

Royal Astronomical Society Press Notices
7 April 1997

Supernova 1987A 10 years on: a review

By Dr Peter Meikle, Imperial College

[Dr Peter Meikle of Imperial College, London, opens the National Astronomy Meeting session on Supernovae and Pulsars on Thursday 10th April at the University of Southampton, with a look back at what has happened to Supernova 1987A since it went off on 23rd February 1987, and what astronomers have learned about it from 10 years of observations.]

Supernova 1987A is, by far, the most extensively studied supernova explosion in history. It has provided an astonishing wealth of discoveries about the physical processes which take place before, during and after a core-collapse-induced supernova.

We now believe that SN 1987A arose from the merger of a massive binary star system. The most dramatic evidence for this is the bright circumstellar ring, beautifully imaged by the Hubble Space Telescope (HST). Material that flowed out during the red supergiant phase was subsequently compressed by the blue supergiant wind to form the ring. This, together with the fainter outer loops, was then "lit-up" (ionized) by the ultraviolet flash as the shock of the explosion reached the surface of the star. While the progenitor of SN 1987A may have been unusual, it was not unique. A rather similar system has been recently discovered in our own Galaxy.

That there was a core-collapse in the star was firmly, and uniquely established by the detection of 19 neutrinos on 23 February 1987. It seems likely that the collapse resulted initially in a neutron star at the centre. However, there is still no convincing evidence for a pulsar at the centre. Indeed, the possibility that the core subsequently collapsed to form a black hole is still being considered. Understanding is growing as to how the core-collapse ultimately leads to the powerful ejection of most of the star. The key lies in the neutrinos, which emerge from the neutron star following the core-collapse. These particles can deposit a large amount of energy in the material just above the neutron star, producing instabilities and coupling the released gravitational energy to the star's outer layers. Observational evidence supporting this scenario has come from infrared spectra of SN1987A. However, it is still not understood what determines the fraction of neutrino energy that is actually deposited.

As the shock travelled through the star, explosive nucleosynthesis took place, creating new chemical elements. In particular, about 0.1 solar masses of nickel-56, which is radioactive, was formed. Several pieces of new evidence for this came from SN1987A. Of particular note was the demonstration, using infrared spectroscopy, of the presence of the decay product cobalt-56 and its decay, in turn, to iron-56. It is the decay of these and other radioactive elements which are responsible for maintaining the emission from the supernova. It is now generally accepted that the explosion and nucleosynthesis was not nicely spherically symmetric. What appears to have happened is that the nickel-56 was created in, perhaps 100, dense, localised pockets, or possibly even finger-like structures pointing radially outward. As the 56Ni decayed first to cobalt-56 and then to iron-56, these regions expanded relative to the rest of the ejecta. Extensive large-scale mixing occurred. Again, much of this is based on spectroscopic studies, especially at near- and far-IR wavelengths. (Asymmetric explosions could easily account for the high velocities many pulsars are observed to have.)

As the ejecta expanded, we saw not just atoms, but also the formation of substantial quantities of molecules and dust. Indeed, carbon monoxide played an important role in cooling large parts of the ejecta. The dust formed in dense, opaque fingers, which seem to be still opaque even after 10 years. As the supernova aged beyond about two years, two important nebular effects took place. These are known as the 'infrared-catastrophe' and the 'freeze-out', and were seen for the first time in SN1987A. The recognition of these processes is vital to our understanding of the way the light curves and spectra of supernovae develop long after the initial explosion.

At the moment, the emission from the ejecta of SN1987A is considerably fainter than from the circumstellar material which dominates the appearance of the SN 1987A system. Nevertheless, the HST has recently shown clearly that the ejecta have "split" into two blobs moving in opposite directions, confirming both the early indications and also the more indirect evidence for an asymmetric explosion. A particularly intriguing fact is that the line joining the two blobs lies exactly along the line to the unexplained and much-disputed Mystery Spot seen briefly a couple of months after the explosion in 1987.

As early as 1990, radio observations began to show the impact of the ejecta with the circumstellar material. The radio emission is steadily brightening, and the effects of the interaction have now also been detected in X-rays and the far-infrared. By about 2006, the ejecta will collide with the ring, with dramatic results. We should see the luminosity increase by a factor of 100 to 1000. This event will provide a beautiful "experiment" for a real-time study of shock interaction and particle acceleration.


Goddard Space Flight Center, Greenbelt, MD
Space Telescope Science Institute, Baltimore, MD


Though the brightest supernova in four centuries lit up the southern sky almost exactly 10 years ago on Feb. 23, 1987, astronomers have waited a decade for the ballooning fireball to become large enough -- about one-sixth of a light-year -- to be resolved from Earth's orbit with NASA's Hubble Space Telescope (HST).

Astronomers announced today that a close monitoring of the supernova (designated SN1987A) with HST's sharp view has resolved a one-tenth light-year long dumbbell-shaped structure consisting of two blobs of debris expanding apart at nearly 6 million miles per hour from each other.

"This structure is a bit of a surprise," says Jason Pun of Goddard Space Flight Center, Greenbelt, Maryland. "This is the first time we can see the geometry of the explosion and relate it to the geometry of the large glowing ring system around the supernova, which has an hourglass shape. The images may yield important clues to the dynamics of the supernova explosion and the structure of the progenitor star."

Pun says the dim area between the blobs may be related to the equatorial belt of material seen around the supernova that existed before the star exploded. The ring was illuminated by the supernova in 1987 during the explosion and has been slowly fading since then.

Ever since the star self-destructed in 1987 astronomers realized that it offered a once-in-a-lifetime possibility, because of its close proximity to Earth, to obtain images of the explosion at various stages and look for any changes in the shape and dynamics.

The latest findings are the result of the Supernova Intensive Study collaboration, headed by Professor Robert Kirshner of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts. Images of SN1987A were taken in September 1994, March 1995, and February 1996 with the Wide Field and Planetary Camera 2 (WFPC2). These results are being presented today at the 189th Meeting of the American Astronomical Society in Toronto, Canada, by co-investigator Pun.

The explosion of the supernova debris appears to be perpendicular to the plane of the inner ring. This suggests that whatever properties that the pre-supernova star has, such as rotation or the existence of a companion star, that is responsible for the formation of the inner ring, may also have influenced the dynamics of the explosion.

The explosion was triggered 10 years ago when the collapse of the star's core sent a blast wave of neutrinos which heated the star's inner layers to 10 billion degrees Fahrenheit. This triggered a shockwave which then ripped the star apart and sent the debris hurtling into space. The fireball has since cooled down (to a few hundred degrees Fahrenheit) and the debris is now heated by nuclear energy from the decay of radioactive nuclei produced in the explosion.

The Space Telescope Imaging Spectrograph (STIS) and Near Infrared Camera and Multi-Object Spectrometer (NICMOS), planned for installation on Hubble this February, will be used to obtain a spatially resolved velocity map of the debris, providing information on the physical conditions of the two blobs.

The debris is expected to collide with the inner ring as early as the year 2002. This will light up all of the dark nebulosity surrounding the supernova, providing new clues to the nature and evolution of the stellar explosion.

Image files in GIF and JPEG format and captions may be accessed on Internet via anonymous ftp from

                                GIF                 JPEG
PRC97-03   SN1987A Fireball     gif/sn87afb.gif     jpeg/sn87afb.jpg

The Space Telescope Science Institute is operated by the Association of Universities for Research in Astronomy, Inc. (AURA) for NASA, under contract with the Goddard Space Flight Center, Greenbelt, Maryland. The Hubble Space Telescope is a project of international cooperation between NASA and the European Space Agency (ESA).



This Hubble Space Telescope picture shows Supernova 1987A and its neighborhood. The series of four panels shows the evolution of the SN 1987A debris from February 1994 to February 1996. Material from the stellar interior was ejected into space during the supernova explosion in February 1987. The explosion debris is expanding at nearly 6 million miles per hour.

Ten years now after the explosion, this cosmic fireball is large enough --- about one-sixth of a light-year in diameter --- to be resolved from the Earth's orbit with the Hubble Space Telescope. The debris is resolved into two opposing blobs and is dim in the center. The apparent direction of ejection is the same as the short axis of the bright inner ring that surrounds the supernova. This suggests that the explosion is directed out of the plane of the ring. The ring is probably composed of materials lost by the pre-supernova star in the last stages of its evolution.

Supernova 1987A is located 167,000 light-years away from Earth in the Large Magellanic Cloud.

The telescope captured the images with the Wide Field and Planetary Camera 2. The central image of the supernova and the ring system was taken in light emitted by nitrogen gas (658 nanometers) on Sept. 24, 1994. The series of debris images were taken using a visible light filter of wavelength around 550 nanometers taken (from left to right) on Feb. 4, 1994, Sept. 24, 1994, March 5, 1995, and Feb. 6, 1996.

Credit: Chun Shing Jason Pun (NASA/GSFC), Robert P. Kirshner (Harvard-Smithsonian Center for Astrophysics), and NASA

Back to ASTRONET's home page
Terug naar ASTRONET's home page