When spectroscopists learned how to determine the surface temperatures of the stars, and the Cepheid variables provided the key to true distances and absolute magnitudes, the way was open for construction of the Hertzsprung-Russell diagram, within which the stars are plotted by absolute magnitude, surface temperature, mass and spectral type.
In the 19th century stars had been classified in alphabetical order in what was thought to be an evolutionary sequence: the true sequence was actually one of mass, from the most massive stars to the least, and proved to be OBAFGKMRNS – “Oh be a fine girl, kiss me right now sweetie.” Our G-type Sun turned out to lie near the centre of the diagram and in the middle of the Main Sequence, a diagonal band running from top left to bottom right, which is occupied by stable stars fusing hydrogen to helium in their cores. The most massive type O stars burn out very rapidly, in under a million years, migrating into the top right of the diagram as red giants, becoming increasingly unstable, and then they collapse so rapidly under their own gravity that they disappear from view into black holes. Type Bs also become unstable red giants, eventually exploding as supernovae and forming neutron stars; type As blow off their outer layers, leaving white dwarfs of condensed matter in the bottom left of the diagram. F and G stars (like the Sun) become orange-red giants and then collapse into white dwarfs; type Ks will follow a less dramatic version of the same course; but the red dwarfs of type M and smaller will last for trillions of years before they finally fade out.
In the 1940s and early 50s it was still thought that all the chemical elements had formed in the Big Bang, although there were major gaps in understanding how. A great achievement of the 1950s, by Hoyle, Alfven, Bondi and others, was to show that everything except hydrogen, helium and small quantities of lithium had apparently formed during the destructive phases of exploding stars, synthesising the heavier elements as they blew apart. When I was a student, we were still taught that these events were the ones seen in the sky as novae, literally ‘new stars’, and one of the big revelations in stellar astronomy ‘since I were a lad’ is that a nova is not an exploding star.
In the 1940s it was thought that they were core explosions of stars and that it might even happen to our Sun, and that was the theme of Arthur C. Clarke’s short story ‘Rescue Party’, reprinted in his collection Reach for Tomorrow (Ballantine, 1956). In the Preface, Arthur wrote, “‘Rescue Party’, which was written in 1945, was my first published story, and a depressing number of people still consider it my best. If this is indeed the case, I have been steadily going downhill for the past ten years, and those who continue to praise this story will understand why my gratitude is so well controlled…” Re-introducing ‘Rescue Party’ in “The Sentinel” (Panther, 1985), in 1985 he wrote, “Those who claim it’s their favourite story get a cooler and cooler reception over the passing years.” It distantly inspired my second published story, ‘Here Comes the Sun’, (Galaxy, March 1971, reprinted in From the Moon to the Stars, 2019), which was set in another planetary system but still which again assumed the event would be a core explosion.
Even that story featured a natural event. But when I.J. Good edited The Scientist Speculates (Heinemann, 1962), Clarke contributed an essay entitled ‘Trouble in Aquila, and Other Astronomical Brainstorms’, in which he wrote, “According to Norton’s Star Atlas, there have been twenty fairly bright novae between 1899 and 1936. No less than five of them have been in one small area of the sky, in the constellation Aquila. There were two in a single year (1936), and the 1918 Nova Aquila was one of the brightest ever recorded. What’s going on in this constellation? Why did 25 per cent of the novae in a forty-year period appear in only 0.25 per cent of the sky? Is the front line moving in our direction?”
Norton’s Star Atlas was one of the set texts for first year astronomy at Glasgow University, so the facts were easy to verify and I marked up the relevant map. As it happened, soon afterwards Professor Sweet covered novae in the astrophysics section of the course, so the ASTRA members in the class buttonholed him at the end of the lecture. He wasn’t impressed by the map, though he conceded that novae probably all had much the same brightness, so the 1918 one would be much closer and the rest would have occurred in a relatively small volume of the Galaxy. Was it simply because they were closer to the Galactic Centre? If so, we might have expected them to be in Sagittarius rather than Aquila. But anyway, how could you blow up a star? And how could the supposed ‘front line’ shift from star to star so rapidly?
I had an answer to that, relating to the lecture he’d just given – which argued that the energy generated by the formation of heavier elements at the core of a pulsating star would eventually become enough to blow them apart. Isaac Asimov’s novel The Currents of Space (1952) turns on the premature ageing of a star’s surface, by infall of heavier elements from interstellar dust. If Clarke’s ‘front line’ was moving so fast, perhaps it involved interstellar travel by space-warp, and perhaps the same technology could be used to place heavier elements such as carbon at the core of a star and age it to the nova phase?
Not surprisingly, Prof. Sweet was less than impressed with the idea of space-warps and matter transmission, but suddenly the idea captivated him. If you could do such a thing, he exclaimed, you’d be better to use simple hydrogen – much more common in the Galaxy, easier to collect, and “if you want to make a fire blaze up, you use petrol, not charcoal!” Seizing the chalk, he began to sketch out on the blackboard the basic equations for blowing up a star, while we watched open-mouthed. But after about six lines he suddenly realised what he was doing, rubbed it all off, tore a strip off me for “unprofitable speculation” and stormed out.
However, we now know that the 1960s analysis of nova physics was invalid. Actually a nova occurs in a binary system, where one star has already collapsed into a white dwarf, and the other has expanded into an orange or red giant whose outer layers are being elongated towards the dwarf star, drawn through the ‘Roche lobe’ into an accretion disc in tight orbit around the dwarf. Once there’s too much material in the disc, there’s a catastrophic collapse of material on to the dwarf’s surface and at that point, because it’s still largely hydrogen from the outer layers of the giant star, it’s compressed enough to generate a surface nuclear fusion explosion. This explains the phenomenon of ‘recurrent novae’, which so puzzled astrophysicists when I was a student, and also why the energy release over time in a nova resembles the profile of a lightning strike. Even Type I supernovae are surface explosions, with the material falling on to a neutron star instead of a white dwarf, and to get a core explosion we have to go to a much more violent Type 2 event, the collapse of a supergiant star. It would make little difference to the inhabitants of planets – if there could be inhabited planets in any such systems, which would be habitable for much shorter periods than our own and already long past that stage by the time of the explosions.
In February 2006 the recurrent nova RS Ophiuci exploded, providing unique data from the Swift orbiting observatory as the shockwave expanded from the white dwarf to interact with the stellar wind from the red giant, before expanding into space around the binary system. The shockwave was expanding at 4 million miles per hour, spectacular, but far below the speed of light. The blast from supernova 1987A, in the Large Magellanic Cloud, expanded into space at ten percent of lightspeed, and that was exceptionally fast for even a Type 2 event. (Robert P. Kirshner, ‘Supernova, Death of a Star’, National Geographic Magazine, May 1988.) The outer layers of the star had already blown off 30,000 years before and there were spectacular fireworks when the shockwave caught up with them ten years later.
Almost from the outset, however, there was another object nearby, designated the ‘Mystery Spot’. It was located two light-weeks from the star, and was not visible before the explosion. If it had been expelled by it, it would have travelled at half the speed of light, but there was no suggested mechanism to account for that. Nothing was seen when it was overrun by the shockwaves a few months later and it remains unexplained, but there is one intriguing possibility. Freeman Dyson had suggested that advanced civilisations might deliberately create supernovae to supply materials, and Geoffrey Burbidge had described a way to do it with a 1012 kilowatt gamma-ray laser, using a thousand times more energy than the power consumption of 20th century civilisation (Nature, 1961; Carl Sagan and I.S. Shklovskii, Intelligent Life in the Universe, 1966). Maybe the laser couldn’t handle such energy without being destroyed, and the Mystery Spot was the plasma remnant of the one which triggered SN 1987A.
A supernova would devastate the Galaxy for tens of light-years around, and the late A.T. Lawton was rather appalled at the idea of causing one deliberately. (“Please leave your doors and windows open – we are about to blow the sun out of your sky!”) There could however be a constructive use for ‘star-busting’ technology, in uninhabited parts of a Galaxy. At the time of the ASTRA discussions, it was being suggested that our Solar System might have formed from an interstellar cloud which was collapsed by the interacting shockwaves from two nearby supernovae. Lawton said that if the probability of one such event was tiny, the probability of two coinciding was (tiny) squared! Nevertheless, the evidence is now clear that it’s what happened (J.E. Enever, ‘When Explosions Collide’, Griffith Observer, January 1983). And that raises the possibility that the supernovae were timed – the deliberate productions of galactic gardeners, rather than industrialists.
Later in this course, I’ll discuss the Fermi Paradox, the question ‘Where is everybody?’ Since our Sun is 4.6 billion years old, and there are Population I stars at least twice as old (Alpha Centauri, our nearest neighbour, is an example) – if interstellar travel is possible, why aren’t they here? It only takes one civilisation to go interstellar and, if it can maintain an average expansion at even 1% of the speed of light, it will fill the Galaxy in 20-30 million years. But the Milky Way looks untouched: we don’t even see the ruins of failed high-technology experiments.
Considering the future of high-technology civilisations, Prof. Nikolai Kardashev suggested they fell into three categories. A Type 1 civilisation would control the matter and energy resources of a planet; Type 2, of a Solar System; and Type 3, of a galaxy. The ultimate example of a Type 2 civilisation is often taken to be a Dyson Sphere, whose builders have destroyed their planets in order to build a shell (probably of asteroids) around their star, maximising the use of matter and energy for life-support.
In the discussions which led to the final chapter of my book Man and the Planets, we concluded that a Dyson civilisation was far from in control of its resources. Malthusian population pressure had forced it to use all of its available resources and with nothing left for fresh uses, its situation was highly unstable and it would probably collapse. Significantly, every fictional portrayal of one – starting in 1970 with my own story ‘The Moon of Thin Reality’ and Larry Niven’s novel Ringworld, has portrayed it as collapsed, perhaps intuitively recognising that it’s a bad idea. We concluded that even before reaching Kardashev 2 status, before aspiring towards Kardashev 3, a high-technology civilisation would have to become highly conservationist. If the Galaxy does belong to the gardeners, as above, then the true paradox is that the proof they are out there is that we can’t see them.
As previously mentioned, the oldest stars are now found in the galactic nucleus, the globular clusters and the halo. Confusingly they’re called Population II, because they were discovered after the Population I stars of the disc. Population II stars consist almost entirely of hydrogen and helium, which formed shortly after the Big Bang, and the most massive of them exploded as Type 2 supernovae, seeding the galactic disc with heavier elements which allowed the formation of rocky planets and the evolution of life. In his TV series and book Cosmos, Carl Sagan made much of the fact that we are, lierally, star-dust. But that simple picture has been drastically modifed, only recently, and made surprisingly little news. It began with the sixth gravitational wave event ever to be detected, by the LIGO array in the United States, confirmed by the VIRGO array which by then was now online in Italy.
Back in the mid-1970s, the USA orbited a series of satellites called VELA which were designed to detect the gamma-rays from clandestine nuclear tests. Whether they did in fact detect an Israeli one, conducted under cover of a storm in collaboration with the South African Navy, or vice versa, remains a contentious issue to this day. But what they did detect, confirmed by the Mariner 10 probe then en route to Mercury, were gamma-ray bursts frequently reaching Earth from events much more distant in space. Just what those were, has remained a source of speculation ever since, although suspicion has increasingly focused on neutron stars, and recently, once pairs of those had been detected, on the possibility of collisions between them.
The first five gravitational wave events detected by LIGO were all collisions between pairs of black holes in the mass range about 40 times that of the Sun (sensational enough, because black holes of of that mass hadn’t been thought to exist, let alone in pairs). Where two such superdense, massive objects orbit one another, Einstein’s gravitational theory predicts that they should lose energy through gravitational radiation and spiral together to collide. Neutron star pairs have been found doing just that, at just the rate predicted. The collision would be extremely violent, releasing a burst of gravitational waves as well as electromagnetic radiation in two highly focused beams.
Observatories all over the world have been on alert in hopes to catch such a ‘kilonova’ event or its aftermath, before it fades too far to be observable. On August 17th, 2017, the event now designated GW170817 was detected almost simultaneously by LIGO and VIRGO, while the gamma-ray burst was picked up by the Fermi and Integral satellites. The four observations narrowed the search area to a small enough area of sky for the Swope 1-metre telescope in Chile to find it within hours, pinning it down to the galaxy NGC 4993, 130 million light-years away. The concentrated period of observation which followed gave rise to more scientific papers, in less time, than any previous single event in the history of astronomy. (Ian Steer, ‘GW170817 Update: Surprises from First Gravitational Wave Observed Independently’, Universe Today, 27th October, 2017.)
One of the biggest surprises concerned what was detected in the debris shell surrounding the explosion. There has always been a problem about the formation of the heavier elements above iron and silver in the Periodic Table, even in supernovae, and suddenly it appears that instead they are formed and released by colliding neutron stars. Gold, platinum and many other elements were detected around GW170817 in huge quantities – at least three Earth masses of gold alone – and the infrared and optical observations of the decaying emission confirmed it, with gaps and peaks in the spectrum corresponding to the energy levels at which those elements formed. (Duncan Brown, Professor of Physics, Syracuse University andEdo Berger, Professor of Astronomy, Harvard University, ‘How the Universe Creates Gold’, The Conversation, October 24th 2017, reprinted EarthSky, October 25th).
Although mergers between binary neutron stars are rare, there are enough of them to account for the abundances of those elements in the Milky Way. Studies of dwarf galaxies with fewer stars have found that most of them are lacking those elements, but in the few where neutron stars have occurred, the proportions resemble those in the Milky Way itself.
Page 145 of The Cosmic Mystery Tour, by Nicholas Mee, Oxford University Press, 2019, is headed by a new version of the Periodic Table showing the origin of the elements, with everything above atomic number 41 formed wholly or partly within merging neutron stars. The Cosmic Mystery Tour is a pocket book with small print, especially in the illustrations, and the importance of that diagram is such that it should have at least been given a full page, or a fold-out like the maps in the 1950s Allen & Unwin hardbacks of The Lord of the Rings.
That chart may rank as one of the greatest intellectual achievements of this century, comparable with the Hertzsprung-Russell Diagram in the 20th. Among its many possible implications, one is that there could have been planets like the Earth even earlier in the history of the Galaxy than we supposed. As to where that train of thought may lead, we’ll be looking at that in later articles.
(To be contined).
See also: The Sky Above You – February 2022