The Tortuous Way to Nuclear Fusion

Giuseppe Cima discusses the real perspectives of fusion energy

By Giuseppe Cima

Newspapers make you think that nuclear fusion for electricity production is within reach and that, unlike fission, it is cheaper, cleaner, and safer. Okay, it's not online yet, it is argued, but it's up to us how fast it will supply useful energy. Things appear more complicated than that as soon as we take a look at the extraordinary variety of methods adopted by all the initiatives proposed.

Some use fuel from nuclear warheads, such as Commonwealth Fusion Systems, of which ENI is a shareholder. Others use the boron-proton reaction, such as the TAE, financed by ENEL, Tokamak Energy promotes magnetic confinement, and First Light Fusion of Oxford has been inspired by the "claw" of Alpheus heterochaelis, the gun shrimp claw. There are about thirty companies in this market and just as many, very different, methods. For the most part, these methods have been already studied, and discarded, by the academic community, and yet the industry has not decided which path to follow. This confusion indicates how distant the goal is. Let's revisit the story of fusion and the origin of this explosion of promises.

The inspiration for nuclear fusion came from looking at the sky, it originated with the mystery of the energy that powers the stars. At first they thought of gravity, a non-trivial idea for a body as large as the Sun. Jupiter, for example, a gaseous planet, has a surface temperature twice that of the Earth, powered by gravity while the planet shrinks of a few millimeters per year. Gravity was also Lord Kelvin's hypothesis for the Sun's power during a famous meeting of the Royal Society on January 21, 1887. The surface temperature of the Sun made him deduce that our star had been shining for about twelve million years, much longer than the Bible states.

William Thomson, 1st Baron Kelvin


Paleontologists were upset, gravitational energy doesn't fit the data, fossils on Earth are evidence of a Sun that has been shining for at least several hundred million years. The discussion remained on hold for a while until Arthur Eddington in the 1920s hypothesized that the heat originates from the fusion of hydrogen into helium. It was already known that helium weighs significantly less than four hydrogen atoms and we now know that this difference accounts for the solar radiation. But something was still wrong, the internal temperature seemed too low to support the relevant nuclear reactions. A few years later, new hypotheses emerged on the decay of hydrogen into helium and in Cambridge, in the 1930s the first accelerators experimentally proved the existence of fusion reactions. In the 1940s some illustrious veterans of Los Alamos - the physicists who worked on the atomic bomb - studied the problem, correctly identifying the main reactions and stellar nucleosynthesis was satisfactorily understood. Recent observations of solar neutrinos confirm the hypotheses of the 1940s: fusion of ordinary hydrogen, the same of H2O, provides all the energy radiated by the Sun.

What does this story have to do with the nuclear fusion that is now being talked about so much in the newspapers? Almost nothing. It's surprising but the stars, thanks to their enormous size and mind-blowing pressure, shine at 5,000 degrees while burning exasperatingly slowly, with the power intensity (ratio of power to mass) of a compost pile. If not, they would rapidly explode and disappear. We, on Earth, with our energy range of ten kilowatts per capita, would not know what to do with stellar fusion which has the power of our basal metabolic rate. What could be used on Earth comes from another branch of science: weapons.

After the development of fission nuclear devices in World War 2, we also developed the fusion of light atoms and in the 1950s, in order to overcome the power limits of fission, the first fusion bombs were developed in secrecy. This time hydrogen, the main component of the Sun and of the water molecule, is not the fuel but two of its rare isotopes are. From these isotopes, deuterium and tritium, come the most powerful weapons, such as the mother of all bombs, the Soviet 40 megaton Tsar Bomb (hopefully none is left in stock) that uses a conventional fission bomb to trigger the hydrogen fusion. Cheaper fission weapons, the so called boosted bombs, use the same principle. Now we know what's the ideal fuel for applications of fusion, the one that burns most easily, a mixture of deuterium and tritium, (DT). Almost immediately people started to think of using fusion for electricity production and started research programs, unclassified, from the 1960s.

Magnetic Confinement Fusion, MCF, turned out to be the method that attracted the favor of those looking for continuous combustion of deuterium-tritium and until now, it has remained the preferred way. The most popular MCF device is called TOKAMAK, a Russian acronym attributed to its inventors, Igor Tamm and Andrei Sakharov, (the latter is the same person who developed the H-bomb and civil rights fame- the world of physicists 50 years ago was smaller.)

Fusion is closely linked to thermonuclear weapons: they have in common the raw material, sophisticated and rare components - deuterium and tritium - certainly not just seawater, as we often hear. Deuterium, about .02 % of hydrogen, is readily available and costs only $ 4 per gram. Tritium, on the other hand, is not present in nature, has an average life of ten years, and is produced only by CANDU nuclear reactors. Stocks of tritium in the world consist of about 50 kg accumulated over the years, barely enough for future experiments and it is a thousand times more expensive than gold.

The weapons solved the tritium scarcity by creating it from an isotope of lithium bombarded by their own neutrons. It is the same lithium that is used for modern cell phone batteries and electric cars, but fusion would burn it in such modest quantities as not to increase its scarcity. Nevertheless, for civilian applications of fusion, the possibility of extracting enough tritium from lithium is far from obvious, the theoretical predictions are not good, experiments have never been made, and it is one of the most relevant results we expect from ITER, a gigantic magnetic confinement experiment whose construction in southern France will produce results hopefully in a couple of decades.

It must also be borne in mind that, if fusion ever proved possible, these 1 GW reactors would have to line up by the thousands to supply themselves with the tritium they need to contribute to the 16 TW global energy hunger.

It is already clear that three common statements about fusion, namely that it is near, cheap, and safe, are premature at best. The fuel is not seawater but it is hard to find and is the same stuff as most nuclear devices use, those same weapons which would be so dangerous in the wrong hands.

For now, there are still no fusion weapons, currently, they need a fission trigger, but explosions of pure deuterium-tritium would be very attractive because, at equal destructive power, they are "cleaner", with less radioactive leftover than the plutonium ones. It's no coincidence that the second-largest fusion project in the world, nearly $ 10 billion in funding, the Lawrence Livermore National Laboratory's NIF in California, has been arranged by the Department of Defense and not by the Department of Energy. NIF has shown it can detonate millimeter-diameter deuterium-tritium capsules triggered by the world's largest laser, the size of three football fields. For now they are not explosions of a kiloton, a thousand tons of TNT, but of a milliton, a thousandth of a ton, a stick of TNT. However, we know that, as in all detonations, the difficult operation is to trigger them. NIF is also promoted as a method suitable for a continuous energy production reactor. I leave it to you to imagine how realistic it would be to string hundreds of explosions per minute with sub-millimeter precision for decades, explosions that individually already stress to the limit a steel vacuum vessel the size of a gymnasium.

Inertial Confinement Fusion, ICF, of which the NIF is the best-known example, is not the only alternative to ITER's magnetic confinement, MCF. Between the very high fuel densities of the ICF, hundreds of times the solid, and the very low densities of the MCF, tens of thousands less dense than our atmosphere, intermediate densities could be employed by fusion by exploiting the magnetic field of MCF and the fast compression of ICF. There have been a few attempts in this direction in the history of fusion but at Colleferro, Italy, in the CNEN laboratories, in the 1960s, experiments were carried out with promising results. High-intensity sources of fusion neutrons were produced by imploding magnetized plasmas with the aid of conventional chemical explosives. The reason for the limited popularity of this method as a potential energy source is found again in how difficult it would be to produce high-frequency explosions, for years and years, as would be required in a reactor. Low-powered, low fallout, tactical bombs immediately come to mind as a possibility for the intermediate-density method. From the 1960s onwards there has been little mention of medium density fusion, until a few years ago when a small private initiative was born in Canada: General Fusion (GF).

GF offers something similar to those earlier Colleferro experiments, but with mechanical pistons instead of explosives. According to General Fusion, the new technique would allow the combustion of deuterium-tritium to be repeated cyclically at a potentially attractive rate and cost. In Colleferro, using explosives, they had not even reached ignition and no one would have noted General Fusion, one of the dozens of private fusion initiatives, were it not that Jeff Bezos, of Amazon fame, decided to invest up to two billion dollars in this venture. Regarding the General Fusion proposal, it is a real shame we can't seek the opinion of the CNEN, now ENEA, researchers who carried out the Colleferro experiments. Unfortunately, the signatories of the publications of those times are no longer with us. This observation reminds us of the very long development times of fusion experiments, a crucial problem in this field. Almost equally interesting it would be to ask Jeff Bezos if he ever noticed that the devices whose development he finances could also help to invent new nuclear armaments.

Military applications of civil fusion should certainly be a major drawback, but the inevitable radioactive activation of fusion reactor structures is even more so. Unlike the solid core of a fission reactor, the thin fuel of a magnetic confinement fusion reactor is transparent to the neutrons it produces and they are stopped only by the first solid wall they encounter. Even ITER, just an experiment, will generate tens of thousands of tons of radioactive material whose disposal would certainly contribute to the cost of the kilowatt-hour of a conceptually similar reactor.

The price of energy will ultimately decide the development of nuclear fusion. The comparison, for now, is with natural gas, which produces electricity in the US at the unbeatable cost of 2 ¢ per kilowatt-hour and with the new renewables, wind and solar, now only three or four times more expensive than gas and, in some cases, even less expensive. Nuclear plants are too large to enjoy the economies of scale of mass-produced gas turbines, solar panels and wind turbines. This feature penalizes fusion enormously and there are solid physical and safety reasons to make us think that this situation will not change. Fusion is now too far behind in its industrial development to be able to participate in decarbonization in this century, which is why it must be pursued with a very long-term perspective and kept away from financial speculation.

Giuseppe Cima, I researched nuclear fusion in labs and universities in Europe and the US : Culham labs in UK, ENEA in Frascati, CNR in Milan, University of Texas in Austin and published more than 70 peer reviewed papers in this field. After loosing faith that a deconstructionist approach to fusion could yield better reactor performances than already indicated by present day experiments I started an industrial automation company in Texas. I have now retired in Venezia, Italy, where I pursue my lifetime interests: environmental protection, energy conservation, teaching technology and science, the physics of mechanical horology.