The Telescope in the Ice

by Mark Bowen

  * * *

  On February 23 (twenty-two years to the day after Fred Reines and his collaborators detected the first naturally occurring neutrino in South Africa) an insignificant star in a nearby galaxy burst into the first supernova visible to the naked eye since 1604. The remnant of the earlier event is known as Kepler’s Star, because Johannes Kepler himself followed its course in detail. He had no choice but to observe it by eye, since it preceded the invention of the telescope by about four years.

  No one has ever been lucky enough to be observing a star at the moment it exploded, and such was the case with Supernova 1987a. When the excitement had died down and events were sorted out, it was determined that its first light must have reached Earth in a window of about seventy-eight minutes, the time between the last “non-observation” and the first observation.

  On the evening of February 23, his time, an amateur astronomer named Albert Jones made a routine scan of the sky with a homemade telescope from his driveway in Nelson, New Zealand. Jones knew the sky pretty well. He has been called the greatest amateur astronomer of all time. At his death at the age of ninety-three in 2013, he had made more than half a million variable star observations and received many degrees and awards for his work, including the Order of the British Empire from the queen.

  That night he was checking on some stars in the Large Magellanic Cloud, an irregularly shaped dwarf galaxy, visible only in the southern sky. It orbits the Milky Way at a distance of about 160,000 light years, which makes it one of our nearest neighbors. In galactic terms it’s basically right next door.

  Jones noticed nothing unusual. It was about eleven a.m. universal or Greenwich mean time.

  When he rose at his usual predawn hour to make observations the next morning, he scanned a different portion of the sky since he’d covered the Large Magellanic Cloud the night before. This was a shame. Otherwise he would have been the first person on Earth to observe one of the most spectacular events in our region of the cosmos in 383 years.

  The following night, he was monitoring stars in another portion of the sky when some clouds moved in, so he “poked the telescope” back to his targets of the previous night in the Large Magellanic Cloud. There he “was quite surprised to see a bright stranger.” He quickly called his friend Frank Bateson, director of the variable star section of the Royal Astronomical Society of New Zealand. “Frank,” he said, “there’s a star in the Large Magellanic Cloud where there was no star before.”

  * * *

  Meanwhile, almost five hours before Jones observed the event and a little more than seventeen hours after he made his crucial non-observation (we are now at about 4:20 a.m. universal time on the twenty-fourth), a professional astronomer named Ian Shelton, who was also searching for variable stars, completed a three-hour exposure of the Large Magellanic Cloud with a camera on one of the telescopes at the Las Campanas observatory in Chile. At first he thought the bright spot on his photograph was a flaw on his photographic plate, since it had not appeared in a photograph of the same galaxy he had taken the previous night. After convincing himself that it was real, he walked from his telescope to the control room of one of the observatory’s larger telescopes to discuss the new development with his colleagues. They concluded that an object that bright and that far away had to be a supernova.

  At one point during the conversation, the night assistant on the larger telescope, Oscar Duhalde, chimed in to say that he had seen the object by eye about an hour and a half earlier when he had stepped outside during his coffee break. This would make him the first person who actually saw the event. Duhalde had taken many plates of the Large Magellanic Cloud, so he knew it well. Evidently, he had been intending to tell the astronomers about his sighting when he returned from his break, but one of them was telling an off-color joke when he stepped inside, and by the time it was told and the punch line translated from vernacular English into Duhalde’s Chilean Spanish, he had lost his train of thought. Anyhow, they all walked outside the dome to take a look, and, sure enough, there it was.

  The scientific community, which may be overly precise in such matters, tends to recognize Shelton as the official discoverer of Supernova 1987a. The more generous members of the community give Jones and Duhalde credit as co-discoverers.

  The news traveled quickly. In New Zealand, Bateson phoned Jones’s message to an observatory in Siding Spring, Australia, where everyone dropped what they were doing to focus on the bright stranger. One of the Siding Spring observers, Rob McNaught, then realized that he had included the Large Magellanic Cloud in a wide-field photograph he had taken at 12:20 p.m. universal time on the twenty-third, seventy-eight minutes after Jones had seen nothing. When McNaught developed the negative, the supernova was there. This observation was quite valuable scientifically, as not only did it serve to bracket the time of the event, it also evidenced “the swift rise to brilliance” of the supernova, which strongly constrained the current theoretical models.

  Within a day, virtually every professional and amateur astronomer in the southern hemisphere was marveling at the sight, either through massive instruments on remote mountaintops and plateaus or small store-bought or homemade telescopes in backyards and on other driveways, in every wavelength range from the infrared to the ultraviolet. Over the following months the range was extended to the x– and gamma ray bands, and Supernova 1987a was observed with telescopes and numerous other sorts of detectors, mounted on balloons, rockets, satellites, and at least one airplane.

  Searching their charts, astronomers identified the star that had exploded as a so-called blue supergiant named Sanduleak-69° 202a (after Nicholas Sanduleak who had catalogued it two decades earlier). This was the first supernova to arise from an identified star—and identification brought surprise, because Supernova 1987a turned out to be the most violent type of supernova, a type II, and theorists had previously believed that only red supergiants, which are about ten times larger than the blue variety, could produce that type.

  For many, it was the experience of a lifetime. In a review that appeared in Science about a year later, two leading supernova theorists, Stan Woosley of the University of California at Santa Cruz and Mark Phillips of the Cerro Tololo Observatory in Chile, offered a “personal observation.”

  “Especially during the first few weeks following the supernova, observers and observatories of all nations and from all continents shared data, speculations, and the sheer exhilaration of the moment. Little was held back. In the process some mistakes were made, but were quickly subjected to test and the errors freely admitted and corrected. It was science at its best. The data we have now will occupy theoreticians for at least a decade, but the memory of the shared experience will last even longer.”

  The messenger for all this excitement was the photon, which has delivered pretty much all the news that our ancestors have ever received from the cosmos, since they first raised their eyes to the twinkling stars.

  * * *

  A new cosmic messenger delivered the news of Sanduleak-69° 202a’s death about four hours before the photon did, with exquisitely precise timing, and it told a story from the heart of the star that was inaccessible to the photon. A six-second flash of neutrinos began passing through the IMB detector in Ohio at 7:35:41.37 a.m. universal time on February 23, plus or minus five one-hundredths of a second. The indeterminacy arose from the uncertainty of the clock on the instrument. Kamiokande saw the flash at the same time within its clock uncertainty, which was about one minute. The flash was also observed by the large liquid scintillation detector, similar to the one Cowan and Reines had used to detect neutrinos in the first place, at the Baksan Neutrino Observatory in the Caucasus. All three of these detectors happened to be located in the northern hemisphere. With neutrinos it doesn’t matter.

  The IMB and Kamiokande teams may have been looking for proton decay, but as physicists will, they had kept the door open for neutrinos all along.

  Kamiokande was first out of the gate. February 23 fell on a Mo
nday that year. Their data tapes were sent by bus from Kamioka to Tokyo according to their standard weekly schedule, they were analyzed through the night on Friday, the discovery was revealed to the collaboration at large on Saturday morning—and then to the rest of the world.

  The Americans weren’t as well organized.

  The IMB tapes were in the possession of the M in the collaboration, Michigan, and their leader, Jack Vander Velde, didn’t think the instrument was sensitive enough to see the neutrinos. Most of the Michigan folks also happened to be out of the country that week, attending that year’s Moriond conference at a ski resort in France.

  But John Learned, an active member of IMB even though he was up to his eyeballs in DUMAND, knew they must have seen it. He had written the supernova section of the IMB grant proposal, so he had the numbers at his fingertips. A few days after the event, John cajoled the Michigan group into sending a copy of the tapes to his former student, Bob Svoboda, who had recently moved to a post-doc in Irvine, and he and Svoboda made a plan for scanning the tapes. Not long after he put down the phone from that conversation, John got a call from one of the Japanese scientists, who told him that Kamiokande “had it.” Particle physicists were way ahead of the curve on e-mail, since it was invented at CERN, so John was using it regularly. He shot back a request for the time that they had seen it.

  “As near as I can recall,” he says, “just about simultaneously, I got the time in Hawaii, I called up Svoboda in Irvine, he’s just run the tape, and he says, ‘We got it!’ And I say, ‘Okay Bob, what time?’ And we check the times with each other, and they’re within two seconds. At that moment we knew we had it.…

  “The little amusing inside story is that we then called Michigan to say, ‘Hey, we have it!’ and the first response from Jack Vander Velde was, ‘Oh, bullshit, John.’

  “I said, ‘Fine. You get the tape, and you scan down to event number so and so. Then call me back.’ And so then things went wild.”

  IMB announced their results to the world about ten days after Kamiokande did, and the two collaborations published back-to-back papers in Physical Review Letters on April sixth.

  * * *

  “When all is said and done,” wrote Woosley and Phillips, “the most exciting and unique aspect of SN 1987a will remain the detection of the neutrino burst that signaled the collapse of its iron core into a neutron star. The numbers are awesome.”

  They estimate that the neutrino luminosity during the first second of the star’s collapse was more than four times the photon luminosity of the entire observable universe (“all that matter from which we could have received light since the big bang”). The observable universe is about twenty billion light years across, while this neutrino explosion came from a region thirty miles across. Expressed another way, in that first second, the neutrinos given off by Supernova 1987a carried off about one hundred times the total energy that will be given off by our Sun in its entire ten-billion-year life. “All the nuclear weapons in the world, on the other hand, could only power the sun for a few millionths of a second. Supernovae are by far the most violent events in the universe.”

  Think of the neutrino flash as a thin spherical shell that expanded at about the speed of light. By the time it passed through our planet it had a radius of 160,000 light years (which means that the star actually died that many years ago). Since the surface area of a shell grows by the square of its radius, the intensity of the neutrino flash drops at the same rate. Nevertheless, even this far from the center of the blast, an average of fifty billion supernova neutrinos passed through each square centimeter of our planet each second for between ten and twenty seconds.

  The neutrino being a shy creature, only twenty-four were detected in all. Kamiokande picked up eleven in a span of about thirteen seconds, IMB eight in six seconds, and Baksan five in ten seconds. And luck played a big role: Kamiokande almost missed the flash, because the detector had switched itself into calibration mode, which it did for roughly two minutes out of every hour, just one or two minutes before the neutrinos passed through. And IMB was partially crippled: one of its power supplies had failed several hours before the flash, and a quarter of its light detectors were down.

  The most important thing about these twenty-four neutrinos was that they provided the first direct link between a supernova and the birth of a neutron star. As theorist Adam Burrows put it at a neutrino conference in 1988, “Core collapse has been studied for thirty years and neutron stars for fifty years in blissful theoretical isolation.… These detections provide us with the first definitive tests of the basic theory connecting stellar death, supernovae, and neutron star birth.… Within [about] 10 seconds … that theory was transformed into an astronomy.”

  Not all stars end in supernovae. Lighter stars, such as our Sun, pass quietly into white dwarves. Stars weighing eight or more times the Sun will explode into neutron stars, and above about twenty-five solar masses, they probably leave black holes or exotic and so far unverified creatures named quark stars behind.

  Black holes and quark stars are the only cosmic objects that are believed to be denser than neutron stars. Theory dictates that neutron stars should be about 40 percent more massive than the Sun, and every candidate whose mass has been measured comes in close to that value. They should be only about twenty kilometers across, which would make the force of gravity at the surface more than a hundred billion times stronger than the force we feel on Earth.

  The towering Russian theorist Lev Landau first postulated the existence of neutron stars in 1932, just a few months after James Chadwick discovered the neutron. Two years after that, none other than Walter Baade, who had placed a bet involving a case of champagne with Wolfgang Pauli, joined fellow astronomer Fritz Zwicky in proposing the link to supernovae. “For many decades after these original contributions, and without any observational evidence, the idea of the neutron star was kept alive only by stalwart theorists.” Their existence was finally confirmed in the late 1960s with the discovery of radio pulsars. One of the first pulsars to be discovered was the one at the center of the Crab Nebula, a diffuse, blue, vaguely crab-shaped object in the constellation Taurus, which seemed likely to be a supernova remnant, since the Crab supernova had been observed historically. But this was circumstantial evidence. The neutrinos from Supernova 1987a sealed the deal.

  * * *

  Radiant stars are basically a balancing act between gravity and nuclear fusion. In the core of a young star, the fusion of hydrogen nuclei, that is, protons, into the second-lightest element, helium, generates enough outward pressure to keep the core from collapsing under the pressure of gravity. Thankfully for life on Earth, fusion also produces light and energy.

  Sanduleak-69° 202a was probably about eleven million years old when it died, and the reason for its demise was that it had run out of fuel.

  It had burned hydrogen for about ten of those eleven million years. This is known as the main sequence in the life of a star. When its main sequence ended, Sanduleak-69° 202a consisted of a core of helium, the so-called ash from the hydrogen burning, surrounded by a large spherical envelope of unburned hydrogen, which was now removed from the life-giving production of energy and pressure at the core.

  Stars are the furnaces that produce all the atoms in the universe heavier than hydrogen. Without them, even planets could not form, much less life. We are stardust, as the saying goes.

  The helium burned to produce carbon and oxygen over a span of something less than a million years. At the end of that stage of its life, the star consisted of a core of carbon and oxygen “ash” enveloped by a shell of helium, enveloped in a large cloud of hydrogen.

  And so on and so forth: lighter nuclei fused into heavier nuclei (carbon and oxygen produce neon, sodium, and magnesium), the remaining ash burned into still heavier nuclei, the leftovers retreated to non-participating shells outside the core, and the star took on the structure of an onion: a central, ever shrinking core, surrounded by concentric shells of progressively lig
hter elements. The death spiral also spun faster: each successive stage took less time than the previous.

  The final fuels were silicon and sulfur. It took them only about a week to burn into a core of solid iron. The process stopped there, because iron is a stable nucleus. There is no energetic advantage to burning it; energy is required either to fuse it into heavier nuclei or split it into lighter ones.

  The core of Sanduleak-69° 202a was extremely dense at the moment it was poised to explode. Its diameter was about half the Earth’s, but it weighed just what it had to in order to turn into a neutron star: 40 percent more than the Sun. The entire star, with its concentric shells of lighter elements and its outer envelope of hydrogen (which was blue on account of its temperature), had a diameter of about thirty million kilometers, about one-fifth the distance from the Earth to the Sun.

  With no fusion left to produce outward pressure, the iron core collapsed in about a tenth of a second. The outer surface fell in at about a quarter the speed of light. Neutrinos began to emerge during the collapse, as increased pressure and the breaking up of some iron nuclei promoted inverse beta decay: protons captured electrons, turned into neutrons, and gave off electron neutrinos, which sped away.

  As the core shrank to a diameter of about sixty miles, pressures and densities rose unimaginably and the cataclysm began: nuclei were ripped apart, free electrons combined avidly with protons to produce more electron neutrinos; electron-positron pairs emerged from the high-energy void to provide the leptonic material for more neutrinos and antineutrinos; other, similar reactions produced muon and tau neutrinos; ultimately all flavors of neutrino and antineutrino were produced in roughly equal measure. But the collapsing core was so dense and so opaque that even the neutrinos couldn’t escape. They and the light were trapped inside.

  At the very center of the collapsing star, densities increased by a factor of a million, to a few times that of an atomic nucleus: about 1014 grams per cubic centimeter. This is like packing a pool of water one kilometer square and a hundred meters deep into a cube one centimeter on a side. The nuclear material was packed so tightly that the strong force, which normally binds protons and neutrons together, reversed sign and began pushing them apart. This outward pressure, originating at the very center of the core, presented a “brick wall” to the collapsing matter that was still rushing in, and this material bounced off the wall in a so-called shock wave, which triggered the stellar explosion.