The 4.5-Billion-Degree Furnace

How the fact iron is common on Earth is telling us there must be a furnace out in space at a temperature of at least 4.5 billion degrees

‘What is precious is never to forget

The essential delight of blood drawn from ageless springs

Breaking through rocks in worlds before our earth.’

Stephen Spender

‘Come quickly, I am tasting stars.’

Dom Pérignon (at the moment

of his discovery of champagne)

You walk down the concrete canyon of a city and iron is all around you. It is in buildings which could never climb so high without reinforced skeletons of steel. It is in the railings that border the parks. It is in the cars passing by, the planes flying high above, the train rumbling over the bridge up ahead. Even the blood coursing through your veins and arteries contains iron, the essential component of haemoglobin, which ferries oxygen ceaselessly around your body. Iron is the most abundant element on Earth. It composes 5 per cent of the Earth’s crust and makes up most of the Earth’s core. Perhaps you have never wondered about the ubiquity of iron on Earth. But actually it is telling us something important: that somewhere out in space there must be a furnace at a staggeringly high temperature of 4.5 billion degrees.

The straightforward conclusion to draw from the ubiquity of iron on Earth is that the element-building nuclear reactions inside stars must continue adding nuclear Lego bricks all the way up to iron. Iron is just about the most tightly bound, most stable nucleus in all of nature. It sits at the very bottom of Aston’s valley of nuclear stability. Given the chance, every light nucleus will roll down to the bottom of that valley and transform itself into iron. It can do this by gobbling one light nucleus after another. However, this is not quite as easy as a football rolling down a hill because the electric charge carried by nuclei causes them to repel each other fiercely. Only if this repulsion can be overcome can nuclei stick together and grow bigger. In practice, this requires the nuclei to be slammed together violently, something which happens only at ultra-high temperature inside stars, when nuclei are flying about at ultra-high speed. The bigger nuclei grow, the bigger their electrical charge and the higher the temperature necessary to overcome their mutual aversion. Since iron is a moderately big nucleus – constructed from 56 nuclear building blocks and carrying 26 times the charge of a hydrogen nucleus – its synthesis requires an extremely high temperature. Precisely how high, however, depends on the exact sequence of nuclear reactions involved in building it.

Recall that, in the heart of a red giant star, where the temperature pushes 100 million degrees, helium-4 nuclei slam into each other in threes to forge nuclei of carbon-12.1 If a carbon-12 nucleus, made by this triple-alpha process, is then struck by another helium-4 nucleus, it is transformed into oxygen-16. And this is just the beginning of a long series of nuclear reactions in which a nucleus grows by capturing one helium-4 nucleus, or alpha particle, after another. This ‘alpha process’ can keep going just as long as the temperature inside the star keeps climbing. An oxygen-16 nucleus, by accreting a helium-4 nucleus, is transformed into neon-20. Neon-20, by doing the same, is morphed into magnesium-24. Magnesium-24 into silicon-28. And so on.2

The forging of silicon-28 is of critical importance to the fate of a star. Once a star reaches this stage in its life, the next steps in the alpha process – collectively known as ‘silicon burning’ – race ahead at breakneck speed. Helium nucleus after helium nucleus is added in rapid, rat-a-tat-tat succession. Within a single day, the star arrives at the very brink of making iron – and, as it turns out, total catastrophe. Actually, the addition of the last alpha particle in the long chain actually makes nickel-56, not iron-56. But this nucleus is radioactively unstable and disintegrates quickly into cobalt-56, which then decays into iron-56. Because the last step in the silicon-burning process involves the biggest, most highly charged nuclei, it is this step that determines the temperature needed to forge iron in stars. That temperature is about 4.5 billion degrees. So this is what the ubiquity of iron on Earth is telling us: that somewhere out in the Universe there must be stars whose internal fires burn 300 times hotter than the heart of the Sun.

Attaining such an enormous temperature is a very tall order, and in practice the only stars that can get this hot are monsters more than ten times as massive as the Sun. But the existence of such stars, and such a blisteringly hot furnace, is not enough in itself to explain why iron is so common on Earth. After all, the iron that is forged must somehow get out into space, and it would appear to be securely locked up inside a star.

Exactly how locked-up is evident from the way in which the interiors of stars evolve to the silicon-burning, iron-forging stage. Recall that when a star turns hydrogen into helium in its core, the core becomes helium-rich. Because it is heavier than the exterior of the star, it shrinks and, in the process, heats up. This creates the conditions for helium to fuse to make carbon. So the star now has a helium-burning core surrounded by a hydrogen-burning shell. But as the star turns helium into carbon, the core becomes carbon-rich and heavier than the rest of the star, so it in turn shrinks and heats up. This creates the conditions for carbon to burn. Now the star has a carbon-burning core surrounded by a helium-burning shell surrounded by a hydrogen-burning shell. By now the pattern should be clear. As a star evolves, building up heavier and heavier nuclei, its interior develops a structure reminiscent of an onion. The heaviest elements are made in the core, the next lightest elements in a shell wrapped around this, the next lightest in a shell about that, and so on. By the time a star is turning silicon into iron in its heart, it consists of a large number of element-building onion skins. From the outside in, there are shells of hydrogen, helium, carbon, oxygen, neon … and, finally, iron.

The problem, it would appear, with element-building in stars is that all the elements painstakingly built up over the lifetime of the star are locked away inside. And of all the locked-up elements, none is more locked up than iron, sequestered at the very heart of the onion-like star. The fact that iron is ubiquitous on Earth is therefore not simply telling us that there is a furnace at 4.5 billion degrees somewhere out in space. It is telling us there must also be a way of getting the iron from that furnace out into space so it can enrich the gas clouds floating there – gas clouds which provide the birth material for new stars and planets. That way was discovered at the height of the Second World War by the British astronomer Fred Hoyle.

Like many scientists, Hoyle was assigned war work. In his case it was the development of radar for the ships of the British navy. At the end of 1944, as part of this work, he was sent to a radar conference in Washington DC. When it was over, it had been arranged by the British Embassy that he would fly across the country to San Diego to see a radar unit at the US naval headquarters. Once in southern California he knew he was only a few hundred miles from Mount Wilson Observatory, which boasted the biggest telescope in the world – the 100-inch Hooker telescope. So, in a spare weekend in his itinerary, he took the train north to Los Angeles.

It was at Mount Wilson that Edwin Hubble had made the most important cosmological discovery of the twentieth century: that the Universe is expanding, its constituent galaxies flying apart like pieces of cosmic shrapnel in the aftermath of a titanic explosion – the Big Bang. When Hoyle arrived at Mount Wilson Observatory’s headquarters in Pasadena, most of the astronomers were absent, engaged like their British counterparts in war work. A notable exception was Walter Baade, a German émigré who at the outbreak of war had been classified as an enemy alien and so excluded from military service. This had allowed him to work at the 100-inch, taking advantage of the air-raid-blackened night skies over Los Angeles to see deep into the Universe.3 After Hoyle had spent a weekend on Mount Wilson at the 100-inch, it was Baade who was sent to pick him up and drive him back down to Pasadena. During the journey, the conversation happened to turn to the latest developments in astronomy and, in particular, exploding stars. Hoyle said he thought they were dull, but Baade quickly persuaded him they were anything but.

Baade’s particular interest was in ‘supernovae’, which he had recognised as a distinct class in the 1930s. The key to this recognition was Hubble’s discovery of the true nature of galaxies, which had preceded his discovery that they were fleeing from each other. In 1924, Hubble had used the 100-inch telescope to pick out individual stars in the Great Nebula in Andromeda, thus proving it was not a gaseous nebula within our Milky Way, as many astronomers believed, but an entirely separate island universe of stars – a galaxy – way beyond the edge of the Milky Way. In fact, using a type of star known as a Cepheid variable as a ‘distance indicator’, Hubble was able to determine the distance to Andromeda. Despite being the nearest galaxy, it was so far away that light took millions of years to travel to Earth.

This had implications for exploding stars, or ‘novae’, which Baade had been observing with his Swiss colleague, Fritz Zwicky.4 They had spotted detonations in galaxies. It was suddenly clear that these must be enormously more luminous than the ones observed in the Milky Way. Baade and Zwicky suggested that there must be two distinct classes of exploding stars in the Universe. ‘Supernovae’ were a million times more powerful than ordinary novae, often out-shining an entire galaxy of 100 billion stars.

What could cause such a tremendous conflagration? Where did all the energy come from? Hoyle pondered this question as he travelled to Montreal to catch a flight back to England. He very quickly realised there was only one source of energy big enough: gravitational energy. If a massive star ran out of fuel in its core, it would no longer be able to generate heat to oppose gravity. The core would be crushed mercilessly. In the rapidly shrinking central regions of the star, the temperature would rise to unimaginable levels, causing nuclei to break apart into their constituent protons and neutrons. This catastrophic collapse would even crush all the empty space out of matter, squeezing protons and electrons together to make neutrons. In a matter of seconds, the middle of the star would be transformed into a hard ball of neutrons no bigger than Mount Everest.5 Hoyle did not know how this implosion might be turned into the explosion of a supernova. All he knew was that in the shrinkage of the heart of the star down to a ‘neutron core’, enough energy would be converted into heat to power a supernova. It was gravitational energy again, the same source Kelvin had mistakenly believed was powering the Sun.

Hoyle’s flight home from Montreal was delayed because of bad weather. During his stay in the city, he was surprised to bump into friends from back in Cambridge. They were nuclear physicists and Hoyle knew they had been recruited to a shadowy project called Tube Alloys, which he was fairly sure was a front for the British programme to build a nuclear bomb. The idea of building a bomb had been in the air since late 1938, when, in Berlin, Otto Hahn and Fritz Strassman had announced that a nucleus of uranium, when struck by a neutron, split into two. Since this ‘fission’ spat out more neutrons, which could split more uranium nuclei, it raised the spectre of a runaway ‘chain reaction’ that might unleash an unstoppable tidal wave of nuclear energy. All that was needed to make a bomb of unimaginable ferocity was a large enough lump of uranium.

In 1939, Niels Bohr and John Wheeler had determined that such a runaway chain reaction could be triggered only in a rare type of uranium. Hoyle had assumed that Tube Alloys’ goal was to accumulate a lump of this uranium-235. When this was done, it would send its personnel to North America to test a bomb, far away from the prying eyes of the Germans. However, uranium-235 was chemically identical to the common form of uranium, and separating it out was a formidable task which Hoyle imagined was likely to take years. It seemed impossible to him that Tube Alloys had succeeded so soon. The mystery then was why its personnel were in Montreal.

The only explanation Hoyle could think of was that there must be another, faster route to the bomb, and this must be close to a test explosion. And as a matter of fact, Bohr and Wheeler had indeed found that there was another nucleus, apart from uranium-235, that would undergo fission. Plutonium was not known in nature but might be made artificially from uranium in a nuclear ‘pile’, or reactor. It was the perfect material for a bomb because it was a distinct element and so could easily be separated from uranium.

As Hoyle marked time in Montreal, waiting for the weather to clear, he recalled a rumour he had heard that somewhere in the south-western US a bomb team had been assembled from some of the greatest scientific minds in the free world.6 It seemed odd to him that such a formidable team was needed. He had thought triggering an explosive release of nuclear energy merely required taking two lumps of fissionable material and slamming them together. Once above their ‘critical mass’, a runaway chain reaction would automatically ensue. The existence of the American bomb team indicated to Hoyle that clearly this would not work for plutonium. Somewhere along the plutonium route to a bomb there must be an obstacle. It must explain why Britain had chosen the uranium-235 route that Hoyle had thought was so hard.

So what was the plutonium problem? The only thing Hoyle could think was that when two pieces of the man-made element were slammed together, so much heat was generated by the fissions that it pushed apart the pieces before a runaway chain reaction could take hold, causing a fizzle rather than an explosion. If so, the pieces would have to be forced together. This could be done by surrounding the plutonium with a spherical shell of conventional explosives and imploding it. But Hoyle knew that creating a perfectly spherical blast wave was fantastically hard. This, he finally deduced as he waited for his plane in Montreal, must be the obstacle standing in the way of a plutonium bomb.

At this point, thoughts that had been circling in Hoyle’s head since his meeting with Baade began to come together. It was the word ‘implosion’ that did it. Implosion, he believed, drove the explosion of a supernova. And implosion was evidently what would drive the runaway nuclear reactions of a plutonium bomb. Hoyle put two and two together. Could it be, he wondered, that implosion in a supernova drove the nuclear reactions that built up elements? Could it be that supernovae were the furnaces where iron and all the other elements in our bodies were made?

When the weather cleared, Hoyle flew in a Liberator bomber back to Scotland, high above the U-boat-infested waters of the North Atlantic. Finally, back in Cambridge, he continued to think about the inferno in a supernova as the perfect furnace for building up elements. He realised immediately that in the hell of a stellar implosion, the temperature would rise so incredibly high that every conceivable nuclear reaction would be possible. Nuclei would also be squeezed so close together that those nuclear reactions would occur at a breakneck rate. Such super-fast nuclear reactions create a state called ‘nuclear statistical equilibrium’, in which there is a perfect balance between the processes of creation and destruction. Every nuclear reaction and its opposite runs at exactly the same rate, so each element is built up exactly as fast as it is broken apart. Imagine if water is fed into a tank at exactly the rate it leaks out. Despite the toing and froing, the water level stays precisely the same. Similarly, in nuclear statistical equilibrium the abundance of each element remains unchanged, or ‘freezes out’. Contrary to all expectations, the bewilderingly complex orgy of nuclear reactions has a simple outcome. And crucially, Hoyle realised, that outcome is predictable.

The abundances would depend on only two things: the temperature in the shrinking stellar core, which determined the average energy of motion of the nuclei; and the mass differences between the nuclei, which determined which nuclei would be preferentially made by nuclear reactions. All Hoyle needed was those mass differences. As luck would have it, he bumped into Otto Frisch, the Austrian physicist who in 1939 had first alerted the British government to the danger that the Germans might build an atomic bomb. Frisch had recently returned from the US, where he had been working on the bomb project. He had what Hoyle wanted. From a drawer in his desk, he pulled out a table of nuclear masses which had been compiled by a German physicist called Josef Mattauch.

Hoyle knew that iron and elements of similar mass formed an unusually broad hummock on the abundance mountain slope. It had been christened the ‘iron peak’. It rose from gently sloping foothills in the vicinity of scandium, up through titanium, vanadium, chromium and magnesium to the summit at iron-56, the most abundant of the iron-group elements. On the far side of the summit, the peak plunged steeply down through cobalt and nickel to foothills in the neighbourhood of copper and zinc. Using Mattauch’s table of nuclear masses, Hoyle calculated the frozen-out abundances expected for nuclear statistical equilibrium inside an imploding star. To his amazement and delight he saw that, for a temperature of about 5 billion degrees, those abundances precisely matched the shape of the iron-peak abundances.

It was an epochal moment in the quest to discover the origin of the elements. Hoyle had found the unmistakable fingerprint of element-creation in nature. The discovery convinced him beyond any doubt that stars could indeed become hot enough to forge the elements in our bodies, and that other places, like Gamow’s Big Bang furnace, were not that furnace.7 It would later set him on course to figuring out how the triple-alpha process might make carbon, opening the way to building up all heavy elements inside stars. The big bonus of Hoyle showing that element-building must go on inside supernovae was that a supernova exploded. It would scatter the products of its furnace to the winds of space. There they would later be incorporated into newborn stars and planets, and life like ours.

The supernova story actually goes like this. A massive, highly evolved star eventually develops an onion-like interior, with heavier and heavier elements in shells closer and closer to its core. Finally, it undergoes silicon burning, which creates a core of iron. But this is very bad news for the star. Iron is at the bottom of Aston’s valley of nuclear stability, so when an iron-56 nucleus is struck by a helium nucleus, the result is a nucleus with more energy per nucleon, not less.8 Instead of mass-energy being turned into heat – which has been the case with all the element-building processes in the life of the star up to this point – heat must be turned into mass-energy. The only place that heat energy can come from is the core of the star itself. In other words, heat is sucked vampire-like from the heart of the star. Deprived of heat, the core will be unable to oppose the gravity trying to crush it. It will shrink catastrophically.

And it is in the maelstrom of the imploding core that nuclear reactions of nuclear statistical equilibrium go on. Those nuclear reactions forge the iron-peak elements we find on Earth. They range from titanium to chromium to copper and zinc to iron itself. They are the elements that have made civilisation possible, and for that we have the furnaces of supernovae to thank. The elements ejected into space by supernovae are a combination of nuclei the star has painstakingly built up over its lifetime and the nuclei forged in the hell of the explosion itself. Nobody said nature was simple.

In 1987, the picture of element-building inside an exploding star was put to the test when the first supernova visible to the naked eye in 400 years burst into the sky. It detonated in the Large Magellanic Cloud, a satellite galaxy of our Milky Way, and historic photographic plates showed that the precursor star was a massive sun called Sanduleak –69° 202. The star reached the state where it was burning iron, which sucked the heat out of its core and caused it to implode. Eventually, the temperature in its heart got so high that nuclei came apart into neutrons, which welded together to make a neutron core – a tiny, super-hard ball the size of Mount Everest. The in-falling star literally bounced off the hard surface of this neutron star. This is how implosion was turned into explosion, creating supernova 1987A.

Before the explosion, as the core of the imploding star shrank, the rising temperature not only triggered an orgy of nuclear reactions that created the elements of the iron peak, it also triggered nuclear reactions in the onion-skin layers around the core – layers made mostly of carbon-12, oxygen-16, neon-20 and silicon-28. These became so blisteringly hot that nuclei in them rolled all the way down to the bottom of the valley of nuclear stability. As pointed out before, the most stable nucleus in nature is not iron-56 but nickel-56, which shares the same number of nuclear building blocks. Consequently, it was into nickel-56 that nuclei in the cooler onion-skin layers surrounding the supernova core transformed themselves.9 Nickel-56 spits out a high-energy photon known as a gamma ray, decaying into cobalt-56, with a half-life of six days. In turn, cobalt-56 spits out another gamma-ray photon, decaying into iron-56, with a half-life of 77 days.

The significance of this is that the gamma rays can be detected by gamma-ray observatories in orbit around the Earth. The best supernova 1987A data actually came from the second decay, when cobalt-56 turns itself into iron-56. Astronomers detected these gamma rays. Not only did they have the expected energy but they faded away with a characteristic timescale of 77 days. It was strong enough evidence that the picture of element-building in supernovae is correct.

But there is more.

As the debris of Sanduleak –69° 202 expanded, by rights it should have quickly cooled and faded. However, there was nickel-56 in the debris, and the gamma rays from its decay reheated the debris,10 making the debris glow with visible light. In other words, the gamma rays were responsible for the very light by which astronomers saw the supernova. And that light died away on a characteristic timescale of 77 days. Nickel was turning into iron before the astronomers’ eyes. They were witnessing the clear signature of the formation of iron-56 in a supernova explosion.

The picture that emerged was awe-inspiring. The iron on Earth and in our blood comes from massive stars which detonated as supernovae before the Sun was born.11 Their debris intermingled with the gas and dust floating between the stars. Out of that gas and dust, 4.55 billion years ago, there congealed the Sun and planets. It is a story which is unique in connecting the very big and far away – stars – to the very small and close to home – the atoms of which you and I are composed. Hold up your hand in front of you. You are made of star stuff. As the American astronomer Allan Sandage has remarked: ‘All humans are brothers. We came from the same supernova.’ This is what the abundance of iron on Earth is telling you.

Notes - CHAPTER 6

1. See Chapter 5.

2. Actually, the build-up of elements inside stars is a little more complicated than this. This is because there are often several alternative routes to building up a particular heavy nucleus. For instance, once carbon-12 and oxygen-16 become common in a star, heavy nuclei can be made by their direct fusion. Thus, two nuclei of carbon-12 can stick to make neon-20 plus a nucleus of helium-4. In practice, the direct fusion of carbon-12 and the direct fusion of oxygen-16 can leapfrog many of the nuclei made by the alpha process.

3. Baade’s great discovery was that the stars in the Milky Way fell into two distinct categories. Population I stars, found in the spiral arms of the Galaxy, are dominated by hot, blue stars. Population II stars, in the central ‘bulge’ of the Galaxy, are dominated by cool, red stars. Later, it would become clear that Population I stars are young and therefore dominated by newborn massive stars. Population II stars are old. All the hot, young stars have gone out, so they are dominated by old, red giant stars.

4. Though a brilliant and visionary astronomer, Zwicky was an eccentric and volatile character whose insistence on calling Baade a Nazi, which he wasn’t, eventually led to the shy, quiet Baade living in fear of his life. Zwicky classified people he did not like as bastards or spherical bastards, who were bastards whatever way you looked at them.

5. Every sugar-cube-sized volume would weigh as much as the entire human race. See Chapter 2.

6. The team was at Los Alamos in New Mexico.

7. In fact, the very lightest elements, principally helium, were forged in the Big Bang. The Big Bang model predicts that about 10 per cent of the atoms in the Universe should be helium, forged in the first ten minutes of the Universe’s existence. And this is exactly what is observed.

8. Actually, the alpha process makes iron-58 and nickel-62, both of which buck the trend and have slightly less mass per nucleon than iron-56. But the nucleus made by addition of a helium nucleus is zinc-60, and this does have more mass per nucleon.

9. Instead of 26 protons and 30 neutrons, which is the case for iron-56, nickel-56 contains equal numbers of protons and neutrons – 28 of each.

10. An iron meteorite is a natural alloy of stainless steel that contains nickel-56, chromium, magnesium and cobalt, all of which were synthesised in the furnace of a supernova.

11. How are elements heavier than iron made? Well, we know that big, highly charged nuclei like zirconium and uranium cannot be formed by fusing together smaller nuclei because, even if two nuclei hit each other at close to the speed of light – the cosmic speed limit – it would be insufficient to overcome their mutual electrical repulsion. This leaves only processes in which a nucleus captures neutrons, since they have no electrical charge. However, free neutrons disintegrate in about ten minutes. The only way a nucleus can accrue a lot of neutrons is if it is (a) exposed over a short period (less than ten minutes) to an intense burst of neutrons, or (b) exposed over a long period to a source of neutrons which is constantly replenished. The existence of uranium, which is neutron-rich, requires source (a). Such a source is believed to exist in supernovae when the nuclei in the imploding core come apart into neutrons, prior to making a neutron core. The existence of zirconium, which is not neutron-poor, requires (b). With a lot of time available, nuclei would have had time to beta decay, transforming a neutron in their nucleus into a proton. In fact, such a location exists inside normal but highly evolved stars. But iron nuclei will have to soak up a lot of neutrons to turn into nuclei of zirconium or uranium. This is not likely, which explains why zirconium and uranium are rare on Earth.