Cavendish Laboratory at Cambridge University was a pivotal institution in the early development of nuclear physics. In the 1930s, world leading particle accelerators were built to conduct experiments on nuclear disintegration. One of the contributors to this work was the Australian physicist Mark Oliphant. While working with Ernest Rutherford, he observed nuclei of hydrogen could be made to come together to form heavier nuclei. This was the first experimental evidence for nuclear fusion.
Figure 1- Fusion of deuterium and tritium to produce helium-4 and a neutron.
With Oliphant's discovery as a base, the German-American physicist Hans Bethe (known for passionately opposing nuclear weapons while advocating the peaceful use of nuclear energy) theorised the nuclear reactions which give power to the Sun and all other stars. It came to be known as the proton-proton cycle. Four hydrogen-1 nuclei, superheated at the core of the Sun, smash into each other to form helium-4 throwing out a mixture of other particles and a lot of energy.
To replicate this in the laboratory on Earth is difficult however. All nuclei are positively charged and so repel each other. Getting them to collide requires them to be contained to a small volume while simultaneously being at temperatures and pressures of extraordinary proportions, a concept known in the fusion field as confinement. Only the gravity of 200 trillion trillion tonnes of star is capable of producing this confinement for the fusion of hydrogen-1 (H-H fusion).
Figure 2- The magnetic coils of a tokamak reactors confine the plasma to donut shaped region. (Image courtesy of ORNL)
But with slightly heavier isotopes, such a deuterium, tritium and helium-4, there are other methods open to us. The first, and still the most popular, method of confinement is magnetic confinement. The first experiments were performed in Russia beginning in 1956 at the Kurchatov Institute in Moscow. Scientists led by Lev Artsimovich constructed large donut shaped chambers within a giant magnetic coil. The idea was to introduce matter into the chamber at temperatures so high that the atoms had been stripped of all their electrons, essentially a mass of nuclei, called a plasma. Since all this plasma would be composed of entirely positively charged nuclei, they would follow the lines of the magnetic field generated by the coil and circulate around the toroidal chamber like a race track and not be able to escape.
The Russian scientists gave this design a name: toroidal'naya kamera v magnitnykh katushkakh, which means toroidal chamber in magnetic coils. This long winded name was then turned into the acronym tokamak.
So within the tokamak, the plasma would be at high temperature, the nuclei travelling at high speed and confined to the narrow race track by the magnetic field, theoretically it would be possible some nuclei would be able to smash into each, despite their repulsion and fuse together. In 1968, in Novosibirsk, the T-4 tokamak reactor successfully achieved nuclear fusion in a stationary reactor.
The long road to commercial fusion
In 1978 in Oxfordshire, construction began on the Joint European Torus, the largest tokamak reactor to date. In 1983, experiments began. Research centred on the fusion of deuterium and tritium (D-T fusion), known to be the easiest reaction to sustain. Throughout its decades of operation, it achieved and sustained nuclear fusion for a large fraction of a second. This fusion however produced only 70% of the external energy fed into it the plasma to initiate it in the first place (referred to as a Q value of 0.7).
To improve upon this, an international project named ITER (originally standing for International Thermonuclear Experimental Reactor, but with the full name being dropped for PR reasons) will be constructed in Cadarache, France, aiming to sustain a D-T fusion reactor for eight minutes with a Q value of 10 (10 times as much energy is produced from fusion within the plasma as is put in externally). As part of the move to a more commercially viable fuel cycle, instead of depending on tritium from outside sources, such as fission reactors, tritium is to be bred in situ using lithium-6. Lithium-6 absorbs a neutron to become lithium-7, which decays by alpha emission (more helium-4) to tritium. This particular fuel cycle may be said to be D-6Li although the fusion reaction itself is still D-T.
Experiments in ITER are expected to begin in 2016 and will serve as a springboard to the first commercial scale electricity producing reactors around 2040. Of course, the opposition are already preparing ammunition and this time, their claims are even more outlandish.
Nuclear fusion still produces radioactive waste.
Wastes from nuclear fission are relatively small on an industrial scale. Wastes from nuclear fusion are negligible. The product is harmless helium-4, stable and chemically inert. D-T fusion involves some neutron production, which will cause activation of reactor materials. This means there is potential for medium level waste upon reactor decommissioning, but this is only important at the end of the 50 year life.
Deuterium is only 0.015% of hydrogen. This would make it rather rare and so not a reliable long term source of fuel.
Hydrogen is found in abundance everywhere. A particularly useable source is sea water. A litre of water will contain 15ml of heavy water. If combined with tritium produced from lithium, the third most abundance element in the universe, it will produce 1.3 terajoules, enough to keep a 500MW reactor going for a quarter of an hour. There is a lot of water on this planet and so a lot of heavy water to supply millions of years worth of energy. Both deuterium and tritium can also be produced, and are being produced, artificially from hydrogen-1 through neutron irradiation.
There is no shortage whatsoever of fusion fuel.
The fusion chain reaction is even more powerful than the fission chain reaction. Control could easily be lost.
Confinement is necessary for the fusion reaction to occur and is also the most technically challenging aspect of designing a reactor. Any malfunction will cause a loss of confinement and so a loss of reactivity. It is physically impossible for a fusion reactor to cascade.
The plasma must be kept at temperatures higher than the surface of the Sun. A loss of confinement would be devastating.
This demonstrates a lack of understanding of the difference between temperature and heat. Heat is the amount of energy an object has and temperature describes the thermal behaviour of the object due to that energy. A good way of comparing is to look at the example of a bath tub full of hot water and the flame of a match. The flame of the match will have a higher temperature, maybe a few hundred degrees while the temperature of the bath is at a temperature of maybe 35°C. However, the bath will have much more heat because there is a much larger mass of water than there is exhaust gases from the flame.
While a temperature of 300 million degrees certainly seems very hot, the total mass of fuel in the plasma is less than one gram. This means that the total amount of energy is less than might be expected. If confinement is lost, the plasma will rapidly expand adiabatically and cool extremely fast as the energy is distributed widely.
We were promised nuclear fission was inherently safe as well when it first arrived.
No we were not. Experimental fission reactors such as the EBR had already shown the capacity for a reactor to destruct itself if the reaction were to cascade including one occasion when during a meltdown, a reactor ejected a control rod into a worker's face. The safety of nuclear fission reactors is from choosing the right combinations of fuel, moderator and coolant as well as through defence-in-depth, where multiple barriers protect the outside world from any potential faults that could arise in the core (all of which were lacking at Chernobyl).
Nuclear fusion on the other hand, has never demonstrated it has any such hazards, controllable or otherwise.
Tritium is notoriously difficult to contain. It leeches everywhere.
In a D-6Li reactor, tritium is predominantly produces in situ and consumed equally as quickly. Therefore, there will not be a large build up of tritium.
Fusion fuel is a component of the most powerful nuclear weapons.
The hydrogen bomb uses deuterium and tritium to provide the huge yields that make it the doomsday machine it is. However, to make them fuse in the warhead a very powerful implosion trigger is necessary. This can only be provided by the fission of plutonium-239. A fusion fuel is worthless in a nuclear weapon without the help of fission fuel.
The neutrons from D-T fusion can be used to breed fissile material very rapidly. A fusion reactor could be used as a covert weapons production plant.
A fusion reactor would be a very effective breeder of fissile fuel, far more so than a fast breeder fission reactor. However, if a nation was pretending to run an innocent pure fusion reactor, then they would not be able to bring in fissionable material without immediately drawing suspicion.
Are we going back down the road of "Too cheap to metre"?
The phrase "Too cheap to metre" was the invention of the science writer Lewis Strauss, not of the developing nuclear industry itself. It is often regurgitated by the opposition as proof that the industry promised something they could not deliver, when in fact it was just a utopian dream of an optimistic journalist (a rare commodity these days).
The economics of nuclear fusion are greatly uncertain at this time, not just because it is unclear the cost of operating such a fuel cycle, but also because it is uncertain what the energy situation will be when nuclear fusion matures to the point of commercialisation.
No-one is under any illusions that nuclear fusion will herald some new age of cheap energy. If it is cheap enough, it will be used, if not, it will not. That is down to the energy market to decide.