A Piece of the Sun: The Quest for Fusion Energy

CHAPTER 1

Why Fusion?

WE OWE EVERYTHING TO FUSION. OUR OWN SUN AND every star that shines in the night sky are powered by fusion. Without it, the Cosmos would be dark, cold and lifeless. Fusion fills the Universe with light and heat, and allows life to happen on Earth and probably elsewhere. The Earth itself, the air we breathe and the very stuff we are made of are the products of fusion.

Following the Big Bang, once things had cooled down enough for neutral atoms to form, the Universe was filled with a fairly even distribution of hydrogen, the simplest atom. There was a bit of helium and some of the mysterious dark matter, but the Cosmos appeared to be just hydrogen atoms and empty space. So how did fusion transform this blank canvas into the menagerie of astronomical objects visible today and the ninety-two natural elements we find around us? First it had some help from gravity. Although gravity is a very weak force, over many millennia it acted to pull hydrogen atoms closer together. Clumps of hydrogen formed and as they got bigger they exerted more of a gravitational pull, drawing in more hydrogen.

As these balls of hydrogen grew, the pressure on the gas in the centre of the ball increased because of the weight of all the hydrogen above it, and with this increasing pressure came higher temperature. (Think of inflating a bicycle tyre: the more you pump it up, the hotter it gets.) Higher temperature means that the atoms are moving at higher speeds and in the high-pressure core of a proto-star they collide against each other with increasing violence. At a certain temperature the collisions are so forceful that the atoms’ outer electrons – which have a negative charge – are knocked away from their nuclei which, in the case of hydrogen, are made of just a single subatomic particle, a positively charged proton. The result is a plasma; a hot maelstrom of charged particles.

At high temperatures, plasma – an ionised gas – represents a fourth state of matter after solids, liquids and gases.

(Courtesy of CEA France)

Once the nascent star grows to a certain size – roughly 28,000 times the mass of the Earth – the temperature in its core reaches around 10 million °C and fusion starts. Fusion is simply the melding together of two nuclei to make a larger one but it’s not an easy thing to do because all nuclei, such as the protons knocking around in the core of a star-to-be, have a positive electric charge and similar charges repel each other. When the temperature gets into the millions of °C, however, the nuclei are slamming together with such force that they get past the electric repulsion and are hooked by another short-range force, the one that holds protons and neutrons – their uncharged companions – together in a nucleus. The two colliding protons have to get within a subatomic arm’s length before this attractive force can grab them and bind them together to make a new nucleus. But two protons don’t make a very stable nucleus by themselves, so most of these pairs split apart again almost immediately.

Very, very occasionally one of these brief fusions is quickly followed by one of the protons decaying into a neutron. A nucleus made of a proton and a neutron – known as a deuteron – is very stable and so the new nucleus survives. Over time this process creates more and more deuterons in the heart of the proto-star and once there are enough of them other reactions start to happen. For example, one of the deuterons can fuse with another proton to produce helium-3 (two protons and one neutron) and, once there are enough of them, two helium-3s can fuse to form a helium-4 (two protons and two neutrons) with two protons left over. These reactions form the start of a chain of fusions which eventually also produces the elements lithium and beryllium.

These fusion reactions produce heat as a by-product because a nucleus of helium-3, for example, is slightly less heavy than the pair of reacting nuclei that created it – a deuteron and a proton in this case. This mass isn’t lost; it is converted into energy during the fusion. So once this chain of reactions gets going, and untold numbers of nuclei are fusing, the heart of the proto-star becomes a raging furnace, further raising the temperature and causing more reactions. This simple process transforms the ball of gas into a fully-fledged star and it – or a very similar reaction chain – is what powers all stars, from the very first which are thought to have ignited about 150 million years after the Big Bang and throughout the 13.7 billion-year history of the Universe.

Fusion has more tricks too. Towards the end of a star’s life, when it has burned up all its hydrogen, it starts to consume helium in a reaction chain that can produce beryllium, carbon and oxygen. When all the helium is used up, other chains begin that consume those nuclei to make even heavier ones. In this way, in the dying days of a star, all of the elements up to iron are created by fusion. Finally, when no fusion fuel remains, the remnants of the star collapse under their own gravity. If it is a large star, this collapse will release so much gravitational energy that it would blast the outer layers of the star outwards in a cataclysmic explosion, a supernova. The energy of a supernova is so intense that it causes further fusions in the heavy nuclei remaining in the star’s ash. These fusions produce all the heaviest elements from iron up to uranium and beyond.

So, over the lifetime of a star, fusion takes the raw material hydrogen and forges it into all the other elements of the periodic table. And when the star explodes at its end, it spreads those elements out into space where they mix with fresh hydrogen and then slowly coalesce into new stars and planets. So these second-generation stars and the accompanying planets that form around them contain a mixture of elements, allowing some of the planets to form rocky surfaces, oceans, atmospheres and life. Every atom in your body, apart from the hydrogen, was created by fusion in a long-dead star.

* * *

Scientists spent the second half of the nineteenth century and the early part of the twentieth figuring out what made the Sun and all the stars shine. It was a mystery to them how the Sun could pump out such prodigious amounts of energy for billions of years without running short of fuel. By the late 1930s they had worked out the rough details of the fusion reactions described above and had their answer. That answer planted the seed of an idea into a number of minds. The seed wasn’t able to grow for a while because of World War II but once that was finished it soon began to sprout. What was that seed? It was the idea that if fusion can power the sun for billions of years, could it supply similarly endless energy on Earth, if it could be mastered? The ancient Greek mythological figure Prometheus stole fire from the gods and gave it to humans, leading to progress, technology and civilisation. Could science steal the power of the Sun and rekindle it on Earth for the good of all humankind?

Prometheus came to a sticky end – chained for eternity to a rock for his crime. Postwar scientists didn’t have a vengeful Zeus to worry about and, in fact, thought that taming fusion was going to be relatively easy. Stars make it look easy: lump together enough hydrogen, add gravity and fusion just … happens. On Earth they didn’t have some of the benefits that stars enjoyed, including the weight equivalent to many thousands of Earths pressing down on the core to heat and compress hydrogen to fusion temperatures. Scientists would have to find some other way to heat and compress hydrogen – how hard could it be?

Although the war years had been devastating, they had produced some technological wonders. At the start, some men had still fought on horseback but the war was soon all about fast-moving armoured tanks, long-distance aerial bombardment, vast aircraft carriers and submarines. By the end of the fighting there were rockets able to hit targets hundreds of miles away, planes with jet engines and, ultimately, a bomb able to destroy a whole city. For some scientists after the war there was a sense of optimism. If they could achieve so much in six years of war, imagine what they would be able to do in peacetime.

One of those things was to develop nuclear power. But this was not fusion, it was the other sort of nuclear reaction, fission, the process behind the atomic bombs dropped on Hiroshima and Nagasaki. Fission is, in a sense, the opposite of fusion. In fission some of the very largest nuclei known, such as uranium, are split apart into two new nuclei. The starting nucleus is slightly heavier than the fragments that result from the fission and this missing mass is converted into energy during the process. Unlike fusion, fission doesn’t need high temperature to take place: the large nucleus will split apart easily if hit by a fast-moving neutron. It was the discovery in 1938 that when some nuclei split they produce neutrons as a by-product which led to the realisation that you could start a chain reaction: one nucleus is hit by a neutron; it splits and spits out two neutrons; these hit two more nuclei which split to produce four neutrons, and so on. This chain reaction is what makes an atomic bomb – a fission bomb – possible: if you bring together a lump of one of these so-called fissile materials – such as uranium or plutonium – larger than a certain critical size, the chain reaction will start spontaneously and run out of control, exploding in a split second.

Before the scientists involved in the Manhattan Project – the Allies’ wartime project to develop an atomic bomb – built an explosive, they tested the chain reaction in a controlled way, in the world’s first nuclear reactor. This was built, in secret, at the University of Chicago in a squash court underneath the stands of its sports stadium, Stagg Field. Known as Chicago Pile-1 because it was a pile of uranium and graphite blocks, its construction was supervised by Enrico Fermi, a celebrated Italian-American physicist. Graphite absorbs neutrons and so slows the reaction. The blocks in the pile were carefully arranged so that there was enough uranium placed close enough together to sustain a chain reaction, but not quite enough for it to run away and explode. There was no radiation screening around the reactor and no protection from possible blasts – Fermi was sufficiently confident of his calculations to decide that they were unnecessary. In the mid afternoon of 2nd December, 1942 one of Fermi’s assistants slowly pulled out a graphite control rod from the centre of the reactor. This reduction in the amount of graphite in the reactor was calculated to be just enough to allow the chain reaction to get going. Fermi watched a neutron counter and saw the number of neutrons swell as the rod was extracted. With a group of dignitaries looking on, Fermi ran the first controlled nuclear reaction for almost half an hour and then reinserted the rod to shut it down.

After the end of the war, engineers didn’t waste any time putting this new technology to commercial use. The first nuclear reactor to produce electricity was built in 1951 in the United States. The first to supply power to the grid was in the Soviet Union in 1954. And the first truly commercial nuclear power plant began work in the United Kingdom in 1956. At the time, many expected nuclear power to produce electricity that was cheap and limitless. But there are problems with using fission as an energy source. First, uranium is a finite resource and some predictions say it may get scarce before the end of the twenty-first century. Then there is safety: because fission reactors rely on a chain reaction, it is possible for the reactions to run away too quickly and for the reactor to overheat. A reactor cannot cause a nuclear explosion like a atomic bomb because the fissile material inside is too spread out, but it can overheat and melt the core – as happened at Three Mile Island in 1979 – or catch fire – as at Chernobyl in 1986. Reactors contain many tonnes of uranium or plutonium fuel and, if they have been running for a while, also a lot of radioactive spent fuel, some of which is extremely harmful to people. When accidents happen, the danger is that this radioactive material will spread far and wide. At Three Mile Island the material was contained; at Chernobyl it wasn’t.

Thirdly, there is the problem of waste. A typical nuclear power plant generating 1,000 megawatts (1 GW) of power produces around 300 cubic metres (m³) of low- and intermediate-level waste per year and 30 tonnes of high-level waste. These quantities are tiny compared to the waste – some fairly toxic – from a coal-fired power station of similar output. But nuclear waste, parts of which can remain radioactive for hundreds of thousands if not millions of years, is more of a problem to deal with. Low- and intermediate-level waste can be buried close to the surface and its radioactivity will decline to safe levels in a matter of decades. High-level waste needs to be disposed of in such a way that it will remain inaccessible to humans or any other species for tens of thousands of years. Just imagining how to do that is a tall order and only a few countries have got to grips with the problem by building permanent repositories deep underground that will be sealed when full. Other countries keep their waste in carefully guarded facilities on the surface. All of the world’s reactors combined produce a total of 10,000m³ of high-level waste per year.

In the tumultuous postwar world, as a few countries raced to commercialise nuclear (fission) energy, some scientists realised that it would be a much better idea to try to generate power using nuclear fusion. The arguments in favour of fusion are compelling. First there is the fuel: fusion runs on hydrogen or, more correctly, on two isotopes of hydrogen known as deuterium and tritium. Deuterium is simply a deuteron (a proton and neutron) with an electron added; tritium has two neutrons in its nucleus. If you fuse deuterium and tritium you get helium and a neutron.

Slam them together hard enough and a deuteron (D) and a tritium nucleus (T) will fuse, producing helium (⁴He), a neutron (n) and lots of energy.

(Courtesy of EFDA JET)

Deuterium can be easily extracted from water. One in every 6,700 atoms of hydrogen in seawater is a deuterium atom. That doesn’t seem like much but given the amount of water in the world’s oceans there is enough deuterium there to supply all the world’s energy needs for billions of years. Tritium is trickier because it is an unstable nucleus with a half-life of twelve years and so it would have to be manufactured. The easiest way to do this is with lithium, a metal used in some batteries. Lithium, when bombarded with neutrons, splits into helium and tritium. Any source of neutrons can cause this reaction and, since fusion reactors themselves are prodigious producers of neutrons, it is thought that a portion of a reactor’s neutron output could be devoted to producing tritium fuel for its own consumption. Lithium can be extracted from easily mined minerals and there is enough around to supply the world with power for several hundred years. When that runs out, there is enough lithium in seawater for several million more years.

This seemingly vast oversupply of fuel is understandable when you consider how little fuel a fusion reactor actually needs. A 1-GW coal-fired power station requires 10,000 tonnes of coal – 100 rail wagon loads – every day. By contrast, a fusion power plant with a similar output would have a daily consumption of just 1 kilogram of deuterium-tritium fuel. The lithium from a single laptop battery and the deuterium from 45 litres of water could generate enough electricity using fusion to supply an average UK consumer’s energy needs for thirty years.

Fusion is a nuclear process, so some might be concerned about its safety. There are safety issues associated with fusion, but they are small compared to a fission reactor. It is actually very hard to keep a fusion reaction going and if there is any malfunction in the controls of a fusion reactor the process would naturally just stop. Even if the process did want to run away it couldn’t do so for long because there is very little fusion fuel in the reactor. Unlike a fission reactor, which has years’ worth of fuel in place in its core, the fuel in a fusion reactor at any one time weighs about as much as ten postage stamps and could keep the reactor running for only a few seconds. The fuel stored outside the reactor is at no risk of reacting: it will only start to burn when heated to more than 100 million °C in the reactor.

Tritium is a radioactive gas, so is harmful to people, but as it will be generated on site, a fusion power plant won’t keep a large supply sitting around. In the unlikely event that, for example, terrorists blew up a fusion reactor or crashed a plane into it, or even if an earthquake and tsunami hit the plant as happened at Fukushima, the amount of tritium released would not require any evacuation of nearby residents. In any event, tritium is a form of hydrogen, a buoyant gas once used in balloons and airships; its natural tendency is to drift straight up.

A fusion reactor does produce some radioactive waste but, again, the amount is tiny compared to a fission plant. The ‘ash’ of fusion burning is helium, the harmless inert gas that is used to fill party balloons, lift modern-day airships, and cool NMR machines. The metal and other materials in the structure of a fusion reactor, after decades of being bombarded by high-energy neutrons from the reactions, will become mildly radioactive. So when a plant is dismantled it will need to be buried in shallow pits for a few decades, by which time it would be safe to recycle. There is none of the high-level waste that takes millennia to cool down.

Fusion seems too good to be true and to the fusion pioneers in the late 1940s and early 1950s, although they wouldn’t have known all of these details, it was clear that fusion would be a vastly superior energy source compared to fission. There was a certain idealism to these early followers of fusion, who were almost all in the United Kingdom, the United States and the Soviet Union. All physicists had been shaken by the power unleashed in the Manhattan Project and many felt a sense of responsibility for the devastation that the atomic bombs caused. Fusion provided a way to use nuclear technology peacefully, for the benefit of everyone. Much of the early work on fusion was done in weapons laboratories because that’s where the nuclear physicists were, but many of them left the labs to pursue fusion outside the military complex.

With the technological optimism of the time, the early pioneers expected that they would be able to master fusion in about a decade and then move on to commercial power stations – a similar timeline to fission. They knew that they would have to get their hydrogen fuel very, very hot, at least 100 million °C. At such temperatures solids, liquids and even gases cannot exist, so they would have to deal with plasma – that fourth state of matter that exists in the Sun’s core where negatively-charged electrons and positive ions move around independently. In the middle of the last century scientists didn’t know much about plasmas, especially very hot ones, and they had to come up with a system that could contain the plasma, heat it hotter than the core of the Sun and then hold it there without it touching the sides, because its extreme temperature would burn or melt almost any material.

Undaunted by these hurdles, the early fusion enthusiasts exploited the key difference between plasma and normal gas: that it is made up of charged particles. When charged particles move around in an electric or magnetic field they feel a force pushing them in a particular direction. So researchers started building containers pierced by complicated magnetic and electric fields to push the particles of the plasma towards the centre and away from the walls. Sometimes these were straight tubes, sometimes ring-shaped doughnuts and other shapes. At first they were made of glass – the scientist’s favourite building material – small enough to sit on a lab bench and sprouting an impenetrable tangle of wires, pumps and measuring apparatuses.

Soon researchers worked out how to create plasmas in their devices and how to heat them to high temperature, if only for a fleeting fraction of a second. No fusion yet, but success in containing and heating plasmas encouraged them to try out new ideas, build more devices and build them bigger. Their makers gave them strange names such as pinches, mirror machines, stellarators and tokamaks. One of the reasons they were not getting to fusion temperatures was that too much heat was escaping from the plasma, so they guessed that bigger was better, since it would take the heat longer to escape from the core of the plasma if there was more of it. Soon their machines were too big for lab benches and were taking up whole rooms, then filling large hangar-like buildings.

They encountered other problems too. Using high-speed cameras to observe the plasma – it glows, just as the plasma in fluorescent lights does – they saw it wriggling and bulging as if trying to break free of its bonds. These phenomena, known as instabilities, had never been seen before, perhaps because no one had tried doing these things to plasma before. To work out how to prevent them, the researchers had to adapt or make up the theory of plasmas as they went along.

A pattern began to emerge in fusion research: scientists would build a new machine; when it was working they would make progress towards fusion conditions but not quite as much as they had predicted; this could be because the machine underperformed or they encountered some new unforeseen type of instability; the way forward was to build another bigger and better machine, and so on. Fusion got a reputation for promising a lot but never delivering. The oft-repeated joke was, ‘Fusion is the energy of the future, and always will be.’

By the 1980s the reactors had come a long way from the bench-top devices of the early days. During that decade the biggest fusion reactors built to date were completed: the Joint European Torus or JET, which is the size of a three-storey house, and its US counterpart, the Tokamak Fusion Test Reactor. These were meant to be the reactors that finally made it to the first great milestone of fusion: break-even. This is the situation when the power given off by the fusion reactions is equal to the power used to heat up the plasma. Thus far all reactors had been net consumers of energy. If fusion was going to be viable as a source of power it had to get over this hurdle. But despite the heroic efforts of researchers over more than a decade, neither of these great machines managed to get to break-even. JET’s best shot, made in 1997, produced 16 megawatts of fusion power but this was only around 70% of the power pumped in to heat the plasma – nearly there, but not quite.

* * *

Some people have spent their whole working lives researching fusion and then retired feeling bitter at what they see as a wasted career. But that hasn’t stopped new recruits joining the effort every year: optimistic young graduates keen to get to grips with a complicated scientific problem that has real implications for the world. Their numbers have been increasing in recent years, perhaps motivated by two factors: there is a new machine under construction, a huge global effort that may finally show that fusion can be a net producer of energy; and the need for fusion has never been greater, considering the twin threats of dwindling oil supplies and climate change.

The new machine is the International Thermonuclear Experimental Reactor, or simply ITER (pronounced ‘eater’) as it now likes to be called. Many machines over the past sixty years have been billed as ‘the one’ that will make the big breakthrough, only to stumble before getting there. But considering how close JET, its direct predecessor, got to break-even, ITER has to have a good chance. Those earlier machines were almost invariably built in haste, part of a breakneck research programme. ITER, in contrast, was in development for a quarter of a century before construction began and, because of the delicate politics of building an international collaboration, was subjected to endless reviews, reappraisals, rethinks and redesigns. It may not be the perfect fusion reactor but it is the best guess of the thousands of researchers who have contributed to its design since the mid 1980s.

ITER is not a power station; it won’t be connected to the grid and won’t even generate any electricity, but its designers are aiming to go far beyond break-even and spark enough fusion reactions to produce ten times as much heat as that pumped in to make it work. To get there is requiring a reactor of epic proportions. The building containing the reactor will be 60m tall and extend 13m underground – altogether taller than the Arc de Triomphe. The reactor inside will weigh 23,000 tonnes – continuing the Parisian theme, that’s more than three Eiffel Towers. The heart of the reactor, the empty space were the hot plasma will hopefully burn, is about four times the height of an adult and has a volume of 840m³ – dwarfing JET’s 100m³.

At the time of writing, workers at the ITER site in Cadarache, in southern France, are laying foundations, erecting buildings, installing cables and generally preparing the ground. In factories around the world the various components that will make up the reactor are being built, ready to be shipped to France and assembled on site. The scale and the quantities are prodigious. In six different ITER member countries factories are churning out niobium-tin superconducting wires for the reactor’s magnets. When finished, they will have made 80,000km of wire, enough to wrap around the equator twice. The giant D-shaped coils of wire that are the electromagnets used to contain the plasma are each 14m tall and weigh 360 tonnes, as much as a fully laden jumbo jet. ITER needs eighteen of these magnets. Perhaps the most mindboggling statistic about ITER, and one of the reasons it is being built by an international collaboration, is its cost: somewhere between €13 billion and €16 billion. That makes it the most expensive science experiment ever built – twice as expensive as the Large Hadron Collider at CERN. The European Union, as the host, is footing 45% of the bill; the rest is being split equally between China, India, Japan, Russia, South Korea and the United States. According to the current schedule, the reactor will be finished in 2019 or 2020.

That huge sum of money is, for the nations involved, a gamble against a future in which access to energy will become an issue of national security. Most agree that oil production is going to decline sharply during this century. There is still plenty of coal around but burning it in large quantities increases the risk of catastrophic climate change. That doesn’t leave many options for the world’s future energy supplies. Conventional nuclear power makes people uneasy for many reasons, including safety, the problems of disposing of waste, nuclear proliferation and terrorism. The disaster at the Fukushima Daiichi nuclear plant in Japan following the earthquake and tsunami in March 2011 served to remind the world how even the most secure installation can still be vulnerable.

Alternative energy sources such as wind, wave and solar power will undoubtedly be a part of our energy future. They are, in a sense, just harvesting energy from our local giant fusion reactor, the Sun. The cost of electricity from alternative sources is high but has declined substantially in recent decades and with continuing improvements in technology it