Alternate View Column AV-23
Keywords: supernova, 1987-A, neutron star, pulsar, neutrino, gravitational collapse
Published in the December-1987 issue of Analog Science Fiction & Fact Magazine;
This column was written and submitted 5/10/87 and is copyrighted © 1987, John G. Cramer. All rights reserved.
No part may be reproduced in any form without the explicit permission of the author.
This page now has an access count of:
Every 50 years or so a star in our galaxy explodes in a supernova. A supernova close enough to the Earth to be seen with the unaided eye was discovered by Johannes Kepler in 1604. In the 383 years since Kepler, astronomers have observed many supernovas in other galaxies, but there has not been one in our own galactic neighborhood close enough to watch in progress.
This long supernova drought was broken early on the morning of February 24, 1987, when Canadian astronomer Ian Shelton developed a photographic plate that he had just taken with a wide-angle 10 inch telescope at the Las Campanas Observatory in Chile. Shelton examined the plate and noticed a new star. An obscure blue B3 supergiant star with the unlikely name of Sanduleak -69o202 had exploded, becoming type II supernova SN1987A. The discovery was broadcast to a data-hungry world, and the astronomy/astrophysics community has been in an uproar ever since. Sanduleak -69o202 before exploding had a mass 15-20 times greater than that of our sun and was located in the Large Magellanic Cloud, a sort of suburb of our galaxy some 160,000 light years distant. To the despair of residents of North America, SN1987A is visible only in the southern hemisphere. A friend just back from Argentina tells me that it's still quite visible there and is about as bright as the stars of the Southern Cross constellation.
What's a type II supernova? It's the explosion of a large star that has burned up all of its nuclear fuel and literally run out of gas. One might think that would produce a dead cinder rather than a giant explosion. But for stars, out of fuel does not mean out of energy. A burned-out star still has a very large amount of gravitational energy. As soon as the nuclear power plant switches off the gravitational energy is "cashed in" as the outer part of the star falls inward.
The star's mantle is no longer held in place by the heat and pressure of nuclear fusion. As it falls in, the density of matter in the interior grows progressively larger. Finally the density rises to about 1013 grams/cm3, about three times the density of an atomic nucleus. At this density, unless the star is so massive that the collapse continues on to form a black hole, the infalling material literally bounces. The density has become so large that the "Pauli repulsion", the intrinsic repulsion of one neutron for another, one proton for another, or one quark for another takes over and halts the collapse of the star.
Then two things happen. Suddenly all the protons in the star's interior are able to gain energy by eating a nearby electron and spitting out a neutrino. In the process they become neutrons, and the inner part of the star becomes a neutron star. While this is going on the outer part of the star is being blown outward by the recoil energy of the bounce. That outflow of energy is now boosted by the additional energy carried to the shock wave by the flood of neutrinos from the core collapse.
Neutrinos are normally very passive particles, electrically neutral and with a mass at or near zero. A neutrino can pass through a light year of lead without scattering from even one lead atoms. Neutrinos usually mind their own business and go their own way without much association with normal matter. But within the passage of a few seconds in the supernova every proton in the interior of the star gives birth to a neutrino. SN1987A probably produced 1058 of them. All of these neutrinos streaming out from the heart of the star must travel through an amount of matter equivalent to a few thousand light years of lead. Therefore, despite their inertness these neutrinos which are carrying away most of the gravitational energy lose about 1% of this energy on their way out. If you were in a Jupiter-type orbit a billion kilometers from SN1987A when it exploded and were protected from the other effects of the supernova, you would be killed by the radiation damage from neutrinos streaming through your body. There are that many. About 100 times more energy from the supernova goes into the neutrinos than into the blast shockwave, and 100 times more energy goes into the shockwave than into visible light. The visible light from the supernova, which can outshine whole galaxies, is a minor side effect of the explosion.
When the outer part of the star lifts off in a giant release of neutrinos, superheated matter and light, it leaves behind a cooling and rapidly spinning neutron star. Why does a neutron star spin? Any star will have some rotation, and in the collapse process it is greatly magnified. Consider an ice skater doing a spin. She begins the turn slowly with arms extended, then draws in her arms to increase the spin rate. The star does the same thing, the matter in collapse playing the role of the skater's indrawn arms. A newly formed neutron star is spun up to such a high rate that it may make a complete rotation in a few thousandths of a second.
Almost a full day before SN1987A was discovered by astronomers, the neutrinos from the explosion reached the earth. Even though the supernova was 160,000 light years away, the number of neutrinos from SN1987A was briefly about equal to the number reaching earth from our own sun. On the average the supernova neutrinos had about 100 times the energy as those from our sun, making them more likely to react with matter. In deep underground locations in the Kamiokande mine in Japan and in the Morton-Thiokol salt mine beneath the shores of Lake Erie are large tanks containing thousands of tons of crystal clear water monitored by thousands of photomultiplier light sensors that wait, up to now without out success, for the light flash that will signal the decay of a proton (see my AV columns, Analog 7/84 and 3/86). On February 23, 1987 these detectors saw no proton decays, but they did see something else. In a space of a few seconds the Kamiokande II detector in Japan recorded 11 neutrino events, and the IMB detector in Ohio recorded 8. The start of the events in the two detectors was essentially simultaneous, within the accuracy of their recording clocks. The Kamiokande and IMB events are the first direct observation of the core collapse of a star. Some of the Kamiokande events were backtracked and found to have come from an approximately 20o cone that included the Large Magellanic Cloud, site of SN1987A. The data tapes containing the neutrino events were not processed until well after Shelton made the visible-light observation of SN1987A, but the neutrinos arrived about 18 hours before there was any visible indication of the explosion. The neutrinos give a prompt signal of the supernova because they are released directly by the formation of the neutron star, while the transport of light and energy through the dense mantle of the star takes many hours longer. In future supernovas the precursor neutrino flash may tell astronomers when and where to look.
In addition to the treasure trove of astrophysical data provided by SN1987A, it can also be viewed as a "experiment" arranged for us by Nature for revealing something about neutrinos themselves. One of the outstanding problems in contemporary physics as been the question of whether, like photons, neutrinos truly have zero rest mass or whether their mass is small but non-zero, as has been suggested by a few questionable experiments and by the dark matter problem (see my AV columns, Analog 2/85 and 5/86). The detection of supernova neutrinos provides a means of weighing the neutrino. If neutrinos have no rest mass then all neutrinos travel at the speed of light and if emitted together should arrive together, even after travelling 160,000 light years. But if neutrinos have non-zero rest mass, those with large kinetic energy will arrive at a detector slightly before those with smaller energy. Since the total spread in the arrival time of the neutrinos was 12 seconds, one can use this to set a limit on the neutrino mass. To do this properly one needs to carefully model the release of neutrinos in a supernova. Adam Burrows of the University of Arizona has done this calculation and gets an upper limit on the neutrino mass of 6 electron volts, with a 99% confidence limit. Other calculations have given larger estimates, but most of these have not included careful modelling of the supernova and the neutrino release process. Burrows' value, if accepted, is quite important because it is smaller than the best upper limits from direct experimental measurements, and it is also smaller by a factor of about three than the mass that a neutrino would have to help with the dark matter problem.
There remain a number of mysteries about SN1987A which are going to provide grist for the astrophysics mill for years to come. First, Sanduleak -69o202 is the wrong kind of star for a supernova. It is a blue supergiant, not an older and larger red supergiant. It is a young hot star that is not supposed to have consumed enough of its fuel to be a supernova candidate. Second, SN1987A seems to be 50 to 100 times dimmer than supernovas previously observed in other galaxies. Is this because Sanduleak -69o202 is a strange and unusual case of a supernova? Or is it just that we have missed similar ones in other galaxies because they were too dim to come to our attention?
But perhaps most mysterious of all is the fact that SN1987A keeps getting brighter. The usual pattern for supernovas in distant galaxies is that the light output begins to dim after a few days, dropping very rapidly for a while and then easing into a more gentle exponential falloff. SN1987A started by following this pattern. But after a few days of dimming, the light intensity began to rise. And it has been rising ever since! From early March to this writing in early May it has shown no signs of decreasing or even of levelling off. SN1987A breaks all the rules.
So what's going on? How can all of these strange characteristics be explained? It's much too early for these problems to be well and truly solved, but let me tell you about a picture that seems to be emerging from the bewildering array of facts. Let's assume that the reason SN1987A seems unusual is because we've never been able to observe such a dim supernova before. It takes a much brighter one to stand out enough to be observed in a distant galaxy. But SN1987A is smaller and dimmer. Since it is smaller, the spinning neutron star in its center plays a larger role as an energy source. Such stars have truly enormous magnetic fields and make a full revolution in a few thousandths of a second. In the process they radiate enormous amounts of electromagnetic energy. This radiated energy, assuming a middling fast rotation rate for the newly born neutron star, is about equal to the amount of visible light now emerging from SN1987A. The neutron star would be superheating the plasma shroud of the supernova from within, the plasma radiating the heat away as visible light. This may be the "power plant" that keeps increasing the light output of SN1987A. If this is so, SN1987A should be a prominent feature of the Southern Hemisphere for some time to come, at least until the neutron star loses some of its juvenile spin and the plasma cloud spreads enough to cool down.
In any case Nature has given us a rare gift, a ringside seat at the metamorphosis of a star from blue giant to neutron star. Astrophysics had begun a new era, and the new science of extra-solar neutrino astronomy has just been born.
"Supernova 1987A: Notes from All Over", M. M. Waldrop, Science 236, 522, (1987).
"Supernova Shines On", R. A. Shorn, Sky & Telescope, 470, (May-1987).
This page was created by John G. Cramer on 7/12/96.