Deuterium and tritium are not the only fuels in fusion; lithium is needed for breeding tritium, which occurs only in minute amounts in nature. Lithium is an abundant element on earth, occurring in two isotopes, 92.6% Li7 and 7.4% Li6. (The super­script is the atomic weight, the total number of protons and neutrons in the nucleus.) Lithium-6 is the more useful one and can easily be enriched to 30-90% for use in a blanket. A 1,000-MW fusion plant will consume 50-150 kg of tritium a year, much more than can be supplied by other sources, such as fission reactors. To generate this amount of tritium in blankets, less than 300 kg of Li6 will be needed by each reactor per year. About 1011 kg of lithium is available on land, and 1014 kg in the oceans. If all the world’s energy is generated by fusion, the lithium will last 30 million years [6]. Deuterium will last even longer. There are 5 x 1016 kg of deuterium in the oceans, and at the rate of 100 kg per reactor per year, that will last 30 billion years! That’s what we mean when we say that fusion is an infinite power source.

The way tritium is made from lithium-6 is shown in Fig. 9.10. The neutron, which started out at 14-MeV energy, has been slowed down by collisions with a moderator material and collides with a lithium nucleus, breaking it into an alpha (a) particle (helium nucleus) and a triton (tritium nucleus). Together, these carry the 4.8 MeVs of energy which is gained in splitting the lithium nucleus. This energy, as well as the neutron’s energy, is transferred to a liquid or gas coolant and eventually transferred to steam for generating electricity. The n-Li7 reaction is the same, except that a slow neutron is left over which can undergo another tritium-producing reaction. The n-Li7 reaction works only with fast incoming neutrons, however.

The problem with this scheme is that not enough tritium is produced, since only 20-40% of the neutrons actually react with the lithium [3]. Some of the neutrons are lost through gaps in the blanket needed for plasma heating and measuring equipment. Some are lost by striking support structures instead of the lithium­bearing material, and a few are lost by passing through the whole blanket. To make up for this, there are fortunately neutron multipliers, mainly lead (Pb208) and beryllium (Be9). These can yield two neutrons for each incoming one. The reaction for beryllium is shown in Fig. 9.11.

Blankets will contain lithium, lead, beryllium, and a structural material; but the main problem is to cool them to take out all the heat that is the power output of the reactor. Blanket designs differ in the way they are cooled and in the form of lithium that is used. To show what is involved, we shall describe three of the leading proposals that have been worked out in detail.