Page 243 - Managing Global Warming

202 Managing Global Warming

Although these fusion fuels appear essentially unlimited, we must be careful not to
be too dependent on other scarce resources of materials also required in the building
and operation of nuclear fusion plants. Potential limitations arise from, for example,
supplies of tantalum (an alloying element in low-activation steels), beryllium, tung-
sten (for plasma-facing surfaces), and helium (itself mostly obtained from natural gas
wells and so currently dependent on the fossil-fuel industry). Estimates of the cost of
fusion power plants also tend to put the use of common building materials such as
concrete and steel at around a third of the total cost, and although this is nuclear-grade
concrete and fusion-specific steel the costs of such basic raw materials can be very
volatile, depending on international economic activity. It is worth remembering
that fusion, like fission, is a capital-intensive energy source and these costs can far
outweigh the costs of scarce materials, which are consumed in small quantities.
There is, however, a remaining fly in the ointment. To start up a fusion reactor will
require at least several kilograms of tritium to sustain power (and neutron production)
until the blanket is producing sufficient supply to maintain self-sufficiency (and also
produce an excess to start other reactors), although this could theoretically be accom-
plished, albeit slowly, by using the neutrons produced from an initial D-D plasma [4].
Current global supplies of tritium are limited to that slowly and expensively produced
in a small number of aging fission reactors [5]. This production could be stepped up if
commercial demand was there, albeit on a multidecade lead time and considerable
forethought would be required. As production fission reactors shut down, the avail-
ability of tritium is forecast to peak and decline following ITER (the next-step inter-
national fusion experiment currently under construction in France); there may only be
enough tritium readily available for one or two power-plant-scale devices following
ITER if they do not breed their own fuel. This bottleneck, and the low rate at which
excess tritium might be produced in a fusion breeder blanket, may substantially limit
the speed with which fusion power plants could be deployed even once the technology
is successfully demonstrated.

5.1.2 Fusion safety

Nuclear fusion will of course require regulation as a nuclear technology. The main risk
to the public arises from the potential for leaks of tritium, and this drives the safety
3
design for conceptual plants. Tritium decays to He through beta emission and as a
hydrogenic species burns to form water, which is easily taken up by biological organ-
isms. In addition, it easily permeates through many materials and systems exposed to
it must be detritiated during decommissioning before they can be recycled. A fusion
power plant may have >10kg of tritium on-site in various systems—and much effort
is devoted to the engineering of systems, which should minimize this.
The design goal for a fusion power plant is zero evacuation—that is, even under
the worst conceivable series of system failures, the amount of released radioactive
material should not be enough to require evacuation of any surrounding area [6]. This,
of course, requires the engineering of multiple levels of safety systems. For example, it
is possible that a plasma disruption could melt part of the interior of the reactor, rup-
turing coolant channels in the blanket and causing the release of tritiated steam at high
pressure into the reactor. Safety engineers model such events and the resulting