eso9001 — Science Release
The Elusive Pulsar in Supernova 1987A
5 January 1990
Is there - or is there not - a pulsar in Supernova 1987A? This is one of the main enigmas in current astrophysical research, and nearly three years after the explosion of the first naked-eye supernova (see eso8704, eso8705, eso8706, eso8711 and eso8802) in four hundred years, the answer is still not known.
In early February 1989, observations of rapid light fluctuations during a single night were reported  tantalizingly hinting at the presence of a blinking pulsar amidst the supernova debris. But, unfortunately, they could not be confirmed and some doubts have been expressed about the reality of these results (eso8902).
However, recent observations by staff astronomers at the European Southern Observatory present new, strong evidence of a pulsar inside the expanding envelope which was blown off by the supernova when it exploded in February 1987. Although the pulsar cannot be seen directly because it is still deeply imbedded in thick dust clouds, its fierce radiation heats these clouds so that they in turn emit strong infrared radiation. The new measurements show that the intensity of this infrared radiation has been constant during the past four months; this is a clear indication of the presence of a central energy source.
The ESO astronomers have alerted the worldwide astronomical community, suggesting that the search for the elusive pulsar should now be intensified .
What is a pulsar?
Current theories predict that the explosion of a heavy star as a supernova will result in most of its mass being blown out into surrounding space, but also that some of it will collapse and be compressed into an extremely dense and rapidly rotating neutron star at the centre. Such an object will later manifest its presence in the supernova by the emission of regular light pulses (hence the name "pulsar"). The pulses are due to the star's very rapid rotation, similar to the intermittent light beam from a lighthouse.
Neutron stars measure no more than 10 - 15 kilometres across, but they weigh as much as our Sun which is about 100,000 times as large. Their densities are correspondingly enormous; one cubic centimetre of neutron star matter weighs one thousand million tons!
Of half a dozen pulsars known within the remnants of old supernovae, the most famous are those in the Crab Nebula and the Vela Nebula, which exploded in the year 1054 and about 10,000 years ago, respectively. The detection of a pulsar inside SN 1987A, the first naked-eye supernova since 1604, would provide definitive confirmation of the creation of pulsars in supernova explosions.
But according to the theory, at early phases the dense gas in the inner regions of the supernova would make it opaque to radiation from a central pulsar and later, the formation of dust clouds would achieve the same effect. Nevertheless, as these clouds expand and become thinner, the light from the presumed new pulsar should eventually penetrate the dust storm and become visible to earthbound observers.
A first glimpse of the new pulsar?
The actual presence of dust in the envelope of SN 1987A was first detected by ESO astronomers in late 1988, by means of accurate observations of emission lines in the supernova spectrum. These dust particles have been created by condensation in the gaseous envelope as it slowly cools. It was found that there is enough dust to hide any pulsar from view and it appeared to be merely a question of time, before the pulsar could be detected through the "thinning smog".
And indeed, in early February 1989 a group of American astronomers reported the discovery of a very fast pulsar in SN 1987A, flashing nearly 2000 times per second. But when observations were performed at ESO a week later, no pulsating light could be seen. This negative result did not necessarily disprove the existence of the purported pulsar; a possible explanation was that the dust around the pulsar was distributed in a clumpy way, and that the pulsar light was therefore being intermittently obscured by dense dust clouds, passing directly in front of the pulsar, as seen from the earth.
Still, the lack of definitive observational proof made many astronomers rather sceptical about the reality of the pulsar.
Infrared observations reveal central energy source
At the present time, almost three years after the explosion, the expanding gas and dust cloud has cooled and is now more than 10,000 times fainter than at the moment of maximum brightness in May 1987. With decreasing temperature, the colour of the cloud has become redder and more than 80 % of its light is now emitted at wavelengths longer than 5 μm, that is in the far-infrared spectral region.
It is therefore necessary to perform infrared photometry (i.e. measuring the intensity of the infrared radiation), in order to evaluate the total energy output from the supernova envelope. Due to the overall faintness of the object and the rather low sensitivity of current infrared detectors, the largest available telescopes are needed for such measurements.
In fact, the only telescope in the southern hemisphere which is equipped to continue such infrared measurements at this time is the ESO 3.6 m telescope. Northern telescopes cannot observe SN 1987A in the Large Magellanic Cloud near the southern celestial pole. The ESO telescope has been used to monitor the supernova almost every month since the explosion.
Three long-time ESO staff astronomers, Patrice Bouchet, John Danziger and Leon Lucy, now report that the amount of infrared radiation from the cloud has levelled off and has in fact been nearly constant since mid-August 1989. A definitive confirmation of this new phenomenon was obtained during several nights in late December 1989, from observations in the 5 – 20 μm spectral region; see the graphic representation of the light-curve, accompanying this Press Release.
These observations show that the temperature of the dust cloud is now 160 K (degrees above absolute zero), or -110°C. The levelling-off of the far-infrared radiation implies that a hitherto-undetected energy source must now be contributing significantly to the total energy output.
Is it a pulsar?
But what is the nature of this energy source? Is it really a pulsar?
One other possibility is radioactive decay of the Cobalt-57 isotope, created during the supernova explosion . However, to account for the infrared intensity now observed, the original amount of Cobalt-57 would have had to be 20 - 25 times larger than the theoretically anticipated amount (about 0.0017 solar mass), and this is in direct contradiction to the observed strength of a spectral emission line from Cobalt ions. Cobalt-57 is therefore excluded as the new energy source.
Could it perhaps be an infrared light echo from dust further out, due to energy emitted earlier from the supernova being absorbed and re-emitted by the dust ? Apparently not, since in that case the radiated light from dust external to the supernova envelope would need coincidentally to have a temperature that closely imitates that of the cooling supernova envelope, and that is highly unlikely. Moreover, a corresponding visible light echo (truly reflected light) is not seen in photometric measurements at shorter wavelengths.
The only other possibility is the powerful emission from a hidden pulsar. Its emitted energy will be absorbed and re-radiated by the dust clouds in the supernova envelope which then shines at nearly constant brightness in infrared light. Weighing all the evidence, the ESO astronomers are convinced that this is the most likely explanation.
They have therefore alerted their colleagues at other observatories to check periodically for the emergence of the pulsar from behind the dust clouds. Nobody knows when that may happen; educated guesses range from a few months to a few years. In addition to rapid light pulses, the pulsar may also manifest its presence by the sudden appearance of certain (high-excitation) emission lines in the supernova spectrum.
The visual magnitude of SN 1987A is now about V = 15.2 and is still decreasing.
 Cobalt-56 was the primary energy source during almost two years, while the brightness of the supernova decreased at a rate that closely followed the half-time life of Cobalt-56 atoms. Cobalt-57 decays 3.5 times slower than Cobalt-56 and would therefore be expected to play a more important role at a later stage.
Tel: +49 89 3200 6276