The Magicians

by Marcus Chown

  From Cambridge University Library, Hoyle borrowed a book written by Victor Goldschmidt. In 1937, the Swiss-Norwegian physicist had carried out a pioneering study of the make-up of the universe, pulling together data from the Earth’s crust, the Sun and from meteorites. The table in which he summarised his results revealed which elements were common and which were rare.

  With Goldschmidt and Mattauch’s data, Hoyle had everything he needed. For a range of different temperatures, he calculated the relative abundances of the elements that would freeze out in nuclear thermodynamic equilibrium, and he discovered something striking: at a temperature of between two and five billion degrees, the relative abundances he predicted for copper and nickel, cobalt and chromium – the elements on which our modern civilisation is based – matched exactly those found by Goldschmidt. Hoyle was euphoric. He now had quantitative proof that such ‘iron group’ elements had been forged in supernovae.5 ‘All humans are brothers,’ as the American astronomer Allan Sandage would one day put it. ‘We came from the same supernova.’

  Hoyle finally had his proof that stars had forged some of nature’s elements. Their interiors were, after all, capable of achieving the enormous temperatures and densities that were necessary. But he believed that it was not just some elements that had been created inside stars but all of them. He was a long way from proving that, but crucially he now had proof that stars could achieve the necessary extreme conditions for ‘nucleosynthesis’.

  The reason people had thought stars could not achieve such conditions – prompting Gamow to look to the Big Bang as an alternative crucible for the building of the elements – was because of an uncharacteristic error made by Arthur Eddington. It was the English astronomer’s detection of light-bending by the gravity of the Sun in 1919 that had simultaneously shown Newton to be wrong and transformed Einstein into a scientific superstar. By the 1930s, astronomers had guessed that starlight was a by-product of the ‘fusion’ of the nuclei of hydrogen into the nuclei of helium.§ However, Eddington believed that the helium ‘ash’ would become mixed throughout a star, gradually diluting its hydrogen fuel and extinguishing the nuclear reactions. Evidence that he might be wrong could be seen in the night sky in stars like Betelgeuse, in the constellation of Orion; such ‘red giants’, far from fading, typically pumped out ten thousand times as much heat as the Sun.

  Making sense of such stars was the problem that had piqued Hoyle’s interest when he had first met Lyttleton. The pair of them had realised that, if a star becomes non-uniform in composition, rather than staying well mixed, as Eddington believed, it automatically becomes hotter and denser, which might explain the light output of a red giant. Hoyle and Lyttleton imagined a star achieving such a non-uniform state by flying through a cloud of interstellar gas and accumulating an outer mantle of hydrogen. This turned out to be unnecessary when Eddington discovered his error.6 The mechanism that he believed mixed helium evenly throughout a star was nowhere near as efficient as expected. Consequently, a star’s helium, being heavier than hydrogen, fell to its centre, where it heated up like any gas that is compressed. As they evolved, stars automatically acquired non-uniform interiors, their cores becoming ever denser and hotter.

  As a star built up heavier and heavier elements, and each fell to its centre, it would develop an internal structure reminiscent of an onion, with each successive layer denser and hotter than the one surrounding it. This was perfect, Hoyle realised, for forging all the elements. When such a star blew up as a supernova or lost matter in a ‘stellar wind’, some of those elements would end up in the interstellar medium as raw material for the next generation of stars.

  In Gamow’s Big Bang furnace, there was only a window of opportunity between about one minute and ten minutes after the birth of the universe in which elements might be forged; after that, cosmic expansion made the fireball too rarefied and too cool. The scheme was able to forge only helium and a few of the very lightest elements. Stellar furnaces, by contrast, had billions of years available for working their alchemical magic. With so much time available, it was patently obvious that stars would win out over the Big Bang. Or would they?

  Gamow’s scheme failed not only because there was less than ten minutes to build up heavy elements in the furnace of the Big Bang but for an even more fundamental reason: in nature, there is no stable nucleus of mass 5 or 8.

  Both protons and neutrons – collectively known as ‘nucleons’ – existed in the fireball of the Big Bang, though neutrons decayed into protons, or hydrogen nuclei, by the time the universe was a little over ten minutes old. A nucleus of the second lightest element, helium, consists of four nucleons – two protons and two neutrons – and so must be built up in several steps. Once helium-4 formed in the Big Bang, the obvious route to building heavier elements was to add another nucleon to make a nucleus of mass 5, or stick together two helium-4 nuclei to make a nucleus of mass 8. But the absence of stable nuclei of mass 5 and 8 in nature meant that the road was blocked, which was as fundamental a problem for the furnaces of stars as for the furnace of the Big Bang.

  After his supernova revelation, Hoyle’s work on element synthesis in stars was therefore stymied, so he turned to cosmology, the science of the large-scale universe. In 1948, together with Hermann Bondi and Tommy Gold, he proposed the ‘steady-state theory’. Edwin Hubble, observing from the Mount Wilson Observatory, had in 1929 discovered that the universe was expanding, its constituent galaxies flying apart like pieces of cosmic shrapnel. According to the steady-state theory, as the galaxies recede from each other, new material fountains into existence in the gaps and congeals to form new galaxies. Although at first sight the idea seems ridiculous, it is no more ridiculous than the idea of all matter erupting into existence in one go in a Big Bang, and it has the advantage that the universe on the large scale looks the same at all times. Such a universe can have existed forever, since only by changing can a universe have an origin. There is no need answer the question: How did it all begin?

  It was partly because of his interest in cosmology that Hoyle attended a meeting of the International Astronomical Union in Rome in the summer of 1952. There, he found himself in the audience of a session on ‘extragalactic nebulae’, or galaxies, being chaired by Walter Baade. The Caltech astronomer had carelessly overlooked the need for a secretary to take the Commission’s minutes, so he asked Hoyle to help out. During the session, Baade presented sensational evidence that the universe was twice as old as had been estimated by Hubble. When, months later, an astronomer who had been in the audience that day stole Baade’s conclusion and passed it off as his own, Hoyle saved the day: his minutes proved that Baade had been outrageously ripped off and ensured that he received the credit he deserved.

  Baade sat on the combined astronomical steering committee of the Mount Wilson Observatory and the California Institute of Technology, which almost certainly explained why, in the autumn of 1952, Hoyle received an invitation to spend three months at Caltech. He jumped at the chance and arrived in Pasadena thinking about nucleosynthesis in stars and possible ways of leapfrogging the troublesome mass 5 and mass 8 gap. Caltech was the perfect place to be; it had both a world-class astronomy department and an active nuclear physics group.

  Research into nuclear physics had begun at Caltech shortly after the construction of the Kellogg Radiation Laboratory in 1930–1. Built with money from Will Keith Kellogg, ‘The Cornflake King’, the laboratory was initially equipped with a powerful 1 MeV X-ray tube, to study not only the physics of such radiation but its application in the treatment of cancer.7 But when John Cockcroft and Ernest Walton sensationally split the atom with high-speed protons in Cambridge, England, in 1932, Charles Lauritsen, the lab’s director, immediately changed the direction of its research.

  An X-ray tube uses a high voltage difference to accelerate electrons so they smash into a metal target, creating high-energy X-rays in the process. It was straightforward to adapt Kellogg’s X-ray tube and use its high voltage difference to instead
accelerate particles such as protons and smash them into atomic nuclei. By observing the resultant shrapnel, the physicists at Kellogg were able to measure the speed of nuclear reactions that transformed one kind of nucleus into another. In fact, when Hans Bethe had proposed the so-called ‘CNO cycle’ of nuclear reactions for turning hydrogen into helium inside stars and generating starlight as a by-product, it was Willy Fowler and his team at Kellogg that had measured the speed of the cycle’s individual nuclear reactions. They discovered that it operated efficiently only at temperatures much higher than the central temperature of the Sun, ruling it out as the principal power source of all but the most massive stars.¶

  Fowler had considered himself merely a nuclear physicist, and it was a revelation when Bethe taught him that what he was doing in the lab might be mimicking the energy-generating nuclear reactions deep inside of stars. Still, it took a young theorist from Cornell University in 1951 to make Fowler realise that his team might also be able to mimic element-building nuclear reactions in stars. Ed Salpeter raised the possibility that the mass 5 and mass 8 barrier might be bypassed by an exceedingly unlikely nuclear process.

  What if three helium nuclei – commonly known as ‘alpha particles’ – came together simultaneously inside a red giant star to create a nucleus of carbon-12? Imagine three people in a supermarket car park crashing their shopping trolleys into each other simultaneously. You would have to wait a long time to see such an event, but one thing stars have in abundance, realised Salpeter, is time – millions or even billions of years, compared with the ten minutes or so in which element-building must be completed in the Big Bang.

  Not surprisingly, Salpeter’s ‘triple-alpha process’ did not work – it was so infrequent that it produced only the tiniest of quantities of carbon. Yet it was an observational fact that carbon is very abundant cosmically – it is the fourth most common element in the universe after hydrogen, helium and oxygen.

  When Hoyle arrived at Caltech at the end of 1952, he was aware of Salpeter’s work, and was sure the Cornell theorist was correct that the only conceivable way to leapfrog the mass 5 and mass 8 barrier was for three helium nuclei to collide and stick together. The question was therefore: Was there any way to speed up Salpeter’s process? Hoyle was certain there was, and he had a wild idea about how to do it.

  Predicting the manner in which the nucleons buzzed about inside a nucleus was beyond the capabilities of any theorist. However, some internal configurations of nucleons were more stable than others, because it was an observable fact that each nucleus could exist in one of a number of ‘energy states’. It had, for instance, a lowest possible energy, or ‘ground’, state, and a number of higher energy, or ‘excited’, states, arranged above it like the rungs of a ladder.

  If there existed an excited state of carbon-12 at precisely the energy of three helium nuclei at the 100-million-degree temperature at the heart of a red giant, Hoyle thought, it would cause the nuclear reaction between three helium nuclei to be ‘resonant’. In exactly the same way that a child’s swing pushed at its natural, or resonant, frequency speeds up, the nuclear reaction would be boosted. Hoyle carried out the relevant calculations and found that, if the triple-alpha reaction to make carbon-12 was resonant, it would be faster than that calculated by Salpeter, not by a factor of ten or one hundred or even one thousand but by an astonishing factor of ten million. Most importantly, Hoyle’s back-of-the-envelope calculations showed that such an enhancement was capable of explaining the abundance of carbon in the universe.

  The energy of three helium nuclei at the 100-million-degree temperature inside a red giant was about 7.65 MeV. For the nuclear reaction to make carbon-12 to be resonant, carbon-12 must therefore have an energy state at precisely 7.65 MeV above its ground state. But did it? This was the question that Hoyle asked Fowler in his office that day in February 1953 – the question which had, fortunately for Hoyle, piqued the interest of Ward Whaling.

  Kellogg Radiation Laboratory, Pasadena, California, February 1953

  Whaling was already an admirer of Hoyle’s chutzpah. On 30 December 1952, shortly after his arrival in Pasadena, the British astronomer had given a public talk on his steady-state theory at the midwinter meeting of the American Physical Society at Caltech. It had created such excitement in the Los Angeles area that it had to be moved to a bigger auditorium at Pasadena Junior College. The talk had so impressed Whaling that he had started attending lectures that Hoyle gave each week in the Robinson Laboratory of Astrophysics, just a few minutes’ walk from Kellogg.

  In his lectures, Hoyle developed his ideas on how element-building might proceed inside stars. Hoyle would invariably be shot down with a killer objection from one of the astronomy department’s big beasts, but it seemed as if criticism from the likes of Jesse Greenstein and Fritz Zwicky was like water off a duck’s back to Hoyle. The following week he would be back with an inventive way of circumventing their objection, only to be shot down again. Whaling found it exhilarating; Hoyle’s knowledge of astronomy and nuclear physics was ropey, but this was compensated for by his exceptional mathematical ability and powerful imagination. Above all, he was eager to learn, and by bouncing back and forth between the nuclear physicists in Kellogg and the astrophysicists in Robinson he learnt fast.

  It was Whaling’s admiration for the bespectacled Yorkshireman that caused him to pipe up and volunteer to look for the excited state of carbon-12. That and the fact that, unlike Fowler, he was not snowed under with other work.

  Whaling’s plan was to use the Kellogg accelerator to fire ‘deuterons’ at nuclei of nitrogen-14. A deuteron is a nucleus of deuterium, or ‘heavy hydrogen’, and contains a proton and a neutron, while a nucleus of nitrogen-14 contains seven protons and seven neutrons. Each collision would result in the creation of carbon-12 and helium-4 nuclei. The key would be to measure the energy of the helium nuclei: the available energy would be shared between the carbon-12 and the helium-4, which would mean that, if the carbon-12 was created in its low-energy, ground state, the helium would be left with a relatively large amount of energy. However, if the carbon-12 was created in a high-energy, excited state, it would leave the helium nuclei with relatively little. Evidence for Hoyle’s predicted state would be the detection of some helium nuclei with precisely 7.65 MeV less energy than the rest.

  As explained in the previous chapter, the energy of a nucleus can be measured by observing how much its trajectory is bent by a strong magnetic field; those nuclei with the highest energy are bent the least and those with the lowest energy the most. A suitably strong magnet was available; the problem was that it was not in the same room as the particle accelerator, and it weighed several tonnes.

  Whaling’s team consisted of his graduate student, Ralph Pixley, a postdoc called Bill Wenzel and a visiting Australian postdoc called Noel Dunbar. None of them could think how to transport the magnet the thirty metres or so down the narrow corridor. Fortunately, their resident engineer, Vic Ehrgott, came up with the ingenious idea of manoeuvring the magnet onto a steel plate resting on hundreds of tennis balls.8 With the weight spread over such a large number of balls, no single ball was crushed flat.

  One member of the team had the job of retrieving tennis balls spat out at the rear of the metal plate and tossing them to the front, where someone else stuffed them under the plate. With a great deal of effort from several red-in-the-face and groaning people, it was possible to inch the load forward. A similar technique using wooden rollers had been used thousands of years before by the workforce of the Egyptian pharaoh to transport blocks of stone from a quarry to the site of the pyramids, and its modern-day incarnation proved equally effective. After two days, Whaling and his colleagues finally had the magnet in the same room as the accelerator and the experiment was ready to begin.

  *

  For ten days, Hoyle was on tenterhooks.9 Each day, he left his office in Robinson and walked the short distance through the winter sunshine to the Kellogg Lab. To his left he could see the tiny
dome of Mount Wilson Observatory high in the San Gabriel Mountains and smell the faint tang of oranges in the air. It was quite a contrast when he plunged into the gloom of the lab and saw Whaling and his team beavering away, surrounded by a jungle of power cables, transformers, whirring vacuum pumps and diving bell-like chambers, in which atomic nuclei were fired at each other.

  Having his prediction tested felt to Hoyle like being in the dock with his life in the balance, while the jury was out deliberating. The difference was that a prisoner knows whether they are innocent or guilty; if they are innocent, they hope the jury gets it right, and if they are guilty, they hope they get it wrong. The jury of experimentalists, however, is always right. ‘The problem is you don’t know whether you’re innocent or guilty, which is what you stand there waiting to hear, as the foreman of the jury gets up to speak,’ said Hoyle.

  On the tenth day, Whaling was waiting for Hoyle. He pumped his hand and gushed his congratulations; the prediction had been borne out. Unbelievably, there was an energy state of the carbon-12 nucleus at 7.68 MeV, which was compatible with 7.65 MeV within the range of experimental error. With the way to bypass the mass 5 and 8 barrier now established, the route to building all heavier elements lay open. Hoyle’s outrageous prediction had been proved right; he had peered into the heart of nature and spied something that mere mortals – or, at least, theoretical nuclear physicists – had been unable to see. ‘The day I heard the result, the scent of orange trees smelled even sweeter,’ he said.10

  ‘It was really quite a tour de force,’ said Fowler. ‘A man walked into our lab and predicted the existence of an excited state of a nucleus, and when the appropriate experiment was performed it was found. No nuclear theorist starting from basic nuclear theory could do that. Hoyle’s prediction was a very striking one.’11