The Physics of Energy, page 84
After this brief introduction we describe the properties of spent nuclear fuel. We then turn to the role of spent fuel in the proliferation of nuclear weapons. Finally we discuss the question of what to do with spent nuclear fuel. Options include reprocessing to reclaim fissile materials and to make the waste less dangerous, permanent sequestration in long-term geological repositories, or managed storage awaiting new technologies and/or political consensus.
Environmental Radiation
On average, human beings living at sea level receive about 3 mSv/y of radiation from natural sources. makes the largest contribution to natural radiation – over 70% in the US – followed by cosmic rays, radioactive sources within our own bodies, and radionuclides other than radon in rocks and soil. Radiation exposure from medical (and dental) diagnostics and treatment and from commercial products vary widely. They average globally around 0.6 mSv/y, but are larger in developed countries, and contribute an additional 3 mSv/y on average in the US. Fallout from nuclear tests, radiation from nuclear accidents, and from fossil fuel combustion currently make negligible contributions to yearly exposure.
20.6.1 The Character of Nuclear Waste
We focus on what is known as high-level nuclear waste, which is basically the spent nuclear fuel (SNF) from nuclear fission reactors. Low- and intermediate-level nuclear waste are much less radioactive. They do not present a severe sequestration problem and do not contain actinides so cannot be used to construct real nuclear weapons. Many schemes and categories exist for classifying nuclear waste, according to its origin, its level of radioactivity, physical state, etc. [124].
Spent nuclear fuel constitutes only ~1% of radioactive waste by volume, but accounts for over 90% of the radioactivity of the waste. The exact composition of SNF depends on the type of reactor and the fuel it uses. CANDU reactors use natural uranium, which is only 0.7% ; most US reactors use low-enriched uranium, with about 3–5% at input; and some, like US nuclear submarine reactors, use highly enriched uranium, with concentration as high as 90%. Some reactors, in France for example, use more complex, partially reprocessed fuel that includes fissile nuclides produced in previous reactor burns. To keep things simple, and because it corresponds to the majority of nuclear waste, we consider the spent fuel from a light-water reactor burning low-enriched uranium [81]. We assume that the reactor is on a four year fuel cycle, i.e. 25% of the fuel is removed every year after spending four years in the reactor. Each year a 1 GWe light-water reactor discharges about 21 tons of SNF (Problem 20.21), containing the principal radioactive species summarized in Table 20.7.
Table 20.7 Characteristic composition of ~21 tons of SNF from a light-water reactor fueled with low-enriched uranium. Abundances are approximate; they depend on the specific configuration of the reactor. All the actinides (U, Pu, Np, Am, Cm) are fissionable with fast neutrons. From [81].
* Fertile Fissile
High probability of spontaneous fission
‖ From β-decay of
The amount of plutonium in SNF is significant. since it is chemically distinct from the other components of SNF, plutonium can rather easily be separated out of SNF. The odd atomic mass isotopes of plutonium, and are fissile. Thus SNF is a source of new fuel for power reactors, but also could provide the raw material needed to make nuclear weapons: 120 kg of would provide about 10% of the fissile material needed for the yearly operation of a 1 GW power reactor. In light of the discussion of uranium reserves in §16, however, a 10% decrease in reactor fuel requirements would not be a decisive factor in the utilization of nuclear power. A reactor designed to breed fuel, like the fast breeders described in §19.4.2, would have an entirely different spent fuel composition. On the other hand, it takes less than 10 kg of to make a nuclear weapon, so the SNF sample summarized in Table 20.7 contains the makings of many fission bombs.
Another significant feature of the composition of SNF is the complexity of its physical and chemical makeup. Fission fragments include materials that are gases or liquids at room temperature and above (xenon and iodine for example); SNF also contains highly reactive elements like cesium and rubidium; and it contains toxic metals like cadmium and, of course, plutonium. These features complicate the handling of SNF, but its radioactivity is the foremost concern.
20.6.2 Time Evolution of Spent Nuclear Fuel
Because it consists of so many different radionuclides, and because many of these give rise to further new radionuclides when they decay, the time dependence of the radioactivity of spent reactor fuel is very complicated. As described in §19.2.2 (see Figure 19.7), the radioactivity and heat generated by nuclear fuel are initially very high, decrease quickly over the first few days, and then decrease more slowly over a period of many years. The IAEA quote an estimate that roughly 10 years after discharge from the reactor, the activity of SNF ranges between and Bq/m3, generating heat at a rate of 2–20 kW/m3 [124].
Spent nuclear fuel retains significant activity from both actinides and fission fragments for thousands, even millions of years. Initially those nuclei with the shortest lifetimes dominate. These are mostly fission products. Then over hundreds of years, actinides, particularly , begin to dominate. The situation switches again around 10 000 years, at which point the long-lived fission fragments such as dominate. Eventually only the longest-lived fission fragments and actinides like uranium isotopes remain. A qualitative graph of the activity of spent fuel and its components is given in Figure 20.20. Since α-decays to , the anomalous radioactivity of nuclear waste in the very long term resembles that of slightly enriched uranium.
Figure 20.20 Time dependence of the activity of one ton of SNF. The line at Bq corresponds to the activity of the amount of uranium ore needed to produce the fuel in the first place. Note that dominates emissions from SNF during the crucial time scale of hundreds to thousands of years. (Credit: International Atomic Energy Agency (IAEA), Nuclear Power, the Environment and Man, IAEA, Vienna (1982).)
From the point of view of environmental stewardship, the intermediate lifetime components of nuclear waste are the most challenging. One can imagine guarding nuclear waste closely for some number of years (either on the site where it was created or in a repository) while the short-lived, very active and very dangerous components decay away. The components with lifetimes of millions of years, though dangerous, have by virtue of their long lifetimes, lower activity levels. The greatest challenge comes from nuclides with lifetimes so long that their safe sequestration cannot be guaranteed, but so short that they have a high level of radioactivity. Actinides like , , and fall into this unhappy middle class.
20.6.3 Spent Nuclear Fuel, Nuclear Weapons, and Nuclear Proliferation
The plutonium isotopes and , which are abundant in SNF, are fissile and could be used in nuclear weapons. Existing nuclear weapons are powered by the fission of or . The critical mass – the minimum mass of fissionable material required for an explosive chain reaction – of an uncompressed sphere of is about 60 kg and for it is only about 10 kg [81]. The power per unit mass of a plutonium bomb is also greater than that of a bomb. Weapons can be made with less fissile material by using various methods to increase the neutron flux within the fissile mass or by dramatically increasing the density by explosive compression. As realistic examples, the first two nuclear bombs used as weapons were made of (1) about 64 kg of enriched to purity, and (2) about 6.2 kg of of very high purity. The first of these weapons was dropped on the Japanese city of Hiroshima, the second on Nagasaki. Both caused massive civilian deaths and destruction.
Note that while there is some concern that low- and intermediate-level nuclear waste (including disused radioactive sources from medical and other uses), could be used to construct dirty bombs that disperse radioactive material through conventional explosives, such devices would be significantly less destructive than full-fledged nuclear weapons, and would not represent nuclear proliferation. While such bombs could be extremely disruptive and careful cleanup efforts could be costly and time-consuming, the actual damage caused would likely not be significantly greater than that produced by the underlying conventional explosive.
and Nuclear Weapons As discussed in §18, is a small component of naturally occurring uranium and must be separated from the chemically identical dominant isotope by physical means. At present it requires a massive industrial operation using complex and sophisticated physical methods to enrich uranium to the high concentrations necessary for nuclear weapons – typically 85% or greater, though a crude weapon can be made at 20% enrichment, or perhaps less. Although several methods have been developed, the dominant one in use at present is centrifugation. The same methods are used to enrich uranium to the ~3–5% concentration suitable for use in a power reactor as are used to enrich uranium to the high concentrations of required for nuclear weapons. It is not easily possible to tell from a distance whether a facility is being used to enrich uranium for peaceful or military use.
Nuclear weapons employing can be conceptually simple and relatively easy to fabricate. In the simplest, gun-type weapon, a mass of highly enriched is assembled in such a way that it is slightly subcritical, i.e. it is too small to sustain a fission chain reaction. At the moment of trigger, an equal or smaller mass of is propelled into the larger making the whole mass supercritical and releasing its energy in a massive explosion as illustrated schematically in Figure 20.21(a). This is how the Hiroshima bomb operated. In a simple weapon like this, the fissioning mass expands rapidly, terminating the fission reaction soon after it starts. It is estimated that only 1% of the in the Hiroshima weapon underwent fission. Modern -based nuclear weapons use neutron reflection and implosion techniques (see below) to reduce the required amount of to 20–30 kg and to increase the explosive yield. The only technologically challenging aspect of constructing a gun-type nuclear weapon is the uranium enrichment. No prototype of this bomb was ever tested during the Manhattan project, since the developers had no doubt that it would work.
Figure 20.21 Schematic sketches of the two dominant designs for fission weapons. (a) A gun-type weapon is technologically unsophisticated and can only be used with . (b) An implosion design can be used either with or plutonium isotopes, and requires high tech shaping and coordination of conventional explosives.
Plutonium and Nuclear Weapons With the exception of minute quantities of , no isotope of plutonium has a lifetime long enough to occur naturally on Earth. The large amount of , as well as smaller quantities of the other fissile isotope , in SNF can, however, be used to fashion nuclear weapons. Unlike , plutonium is both highly radioactive and highly toxic. α-decays to with a half-life of 24 000 y. The α-decay is occasionally accompanied by γ-rays as the nucleus settles down to its ground state. The isotopes and are also α-emitters, and they are more active than because they have shorter lifetimes. In fact, most of the radiation coming from the plutonium in spent power reactor fuel comes initially comes from the . The radioactivity of plutonium isotopes complicates weaponization of plutonium in several ways that make it difficult to construct a powerful plutonium-based nuclear weapon without substantial resources and technical expertise.
The properties of complicate the weaponization of plutonium extracted from SNF. is formed when absorbs a fast neutron without fissioning. Since is not fissile, the ratio of to grows over the period that a fuel element spends in a reactor. By the time the fuel is removed in a normal fueling cycle, may account for ~25% of the plutonium present (Table 20.7). usually decays to by α-emission. It has a probability , however, of undergoing spontaneous fission (see §18.3.2 and Table 18.2). With a half-life of 6561 years, a mole of undergoes spontaneous fission with an activity of Bq (Problem 20.20). When it fissions, produces neutrons that can trigger a premature chain reaction in a plutonium-based nuclear weapon. This makes the gun-type bomb design ineffective even for plutonium with a relatively small content. Unless the critical mass of plutonium is assembled extremely rapidly, premature chain reactions seeded by neutrons from spontaneous fission of leads to a fizzle, in which a small quantity of the plutonium fissions, causing an explosion that scatters the remaining material. Such a device could still cause immense destruction and radiation effects, but would not be as devastating as a device that achieved more complete fission of the . The strategy devised during the Manhattan Project (and basically unchanged to this day) to assemble a critical mass of was to use an array of simultaneous conventional explosions to rapidly compress the plutonium. This type of bomb is called an implosion device and is sketched in Figure 20.21(b). The famous Trinity test at Los Alamos tested this design, which was subsequently used at Nagasaki.
If captures another neutron in a power reactor, it forms , which accounts for about 9% of the SNF described in Table 20.7. is fissile, but its lifetime is short: it β-decays to with a half-life of years. Fresh SNF, which is both thermally and radioactively hot, contains considerable and little . As time goes on the content increases. The short lifetime of complicates its use in nuclear weapons, and the slow buildup of presents a radiation hazard because α-decay is often accompanied by an energetic γ-ray. The activity of is low enough that it can be manipulated using only moderate radiation shielding (radiation protected glove boxes); however, produces times the γ-ray dose of [125]. It must be manipulated remotely if the workers are to be guarded against significant radiation exposure.
The combination of the high degree of radioactivity of SNF from and the technical challenges of building an operational implosion device should make it rather difficult for a small group of people or organization with limited resources to use spent nuclear fuel to develop a fully functional plutonium-based nuclear weapon. A nation-state with full control over uranium-fueled nuclear power plant operation and fuel cycles can, however, circumvent these issues. The IAEA classifies plutonium according to the percentage of , into super-grade (2–3%), military grade (), fuel grade (7–18%), and reactor grade (). Plutonium recovered from power reactors using a normal fuel cycle, in which the fuel elements are removed after 3–4 years produces reactor grade plutonium, which presents all the challenges just mentioned. Super- or military-grade plutonium, however, can be extracted from a reactor designed for nuclear power production if the fuel elements are removed and chemically reprocessed to extract plutonium after a short time – a few months. Then the isotopes and are suppressed and weapons-grade plutonium can be obtained. A sophisticated technical community working under the protection of a national government would then have the time and resources to construct a plutonium-fueled implosion device. This is the approach that has been taken in all plutonium-based nuclear devices developed in the past. By monitoring reactor fuel cycles sufficiently closely, it can also thus be determined whether weapons-grade plutonium is being removed from a functioning reactor.
Thorium and Nuclear Weapons can be used to breed (see §19.3), which can be removed from the reactor and fashioned into a nuclear device. The can be obtained in relatively pure form by chemically separating it from the fuel. Like plutonium, can be used in an implosion-type bomb. Indeed, the US successfully tested a -based weapon in 1955.
The weaponization of seems to be somewhat more difficult than . can easily be denatured by mixing it with natural uranium after separation. The resulting mixture would have too little for weapons use, but enough for reactor fuel. It could not be re-separated without complex isotope separation equipment. On the other hand, the potential utility of highly enriched (which can be used in smaller, more efficient, and more easily controlled reactors) compared to a less concentrated form argues against this. Proliferation resistance for comes from another, less obvious, mechanism [126]: when it is produced from (in one scenario), is accompanied by at an abundance of roughly 0.4%. decays with a 69 year half-life in a decay chain that includes a highly hazardous 2.6 MeV γ-ray that accompanies the β-decay of . Even more than the γ-activity of , this γ-radiation makes the sample too radioactive for easy manipulation. Also, the 2.6 MeV γ-ray, which is hard to shield, makes this material easy to identify. Note, however, that much like the situation with , it is possible to design a reactor configuration and fueling cycle to minimize the contamination of .
Nuclear Weapon Materials
suitable for a nuclear weapon is difficult to separate from natural uranium, but relatively easy to fabricate into a weapon. suitable for a nuclear weapon can be separated by chemical means from spent reactor fuel, but it is more difficult to fabricate into a weapon.
The weaponizability of plutonium from SNF depends on its isotopic composition. The presence of , a neutron source, and which rapidly decays to , a γ-emitter, makes weaponization more difficult.
Plutonium extracted from normal power reactor refueling can be used to create crude but effective nuclear weapons, capable of doing considerable damage. Withdrawing reactor fuel on a more frequent basis yields plutonium with relatively less and and makes weaponization easier.
20.6.4 What to Do With Spent Fuel?
There are (at least) three ways to deal with spent nuclear fuel: first, it can be treated as toxic waste and sequestered from the environment with little or no post-processing; second, it can be reprocessed with the twin goals of (i) reducing the radioactivity and/or toxicity of the waste and (ii) extracting fissile actinides that can be used as additional fuel for power reactors; or third – an intermediate option known as managed storage (see, e.g., [73]) – it can be stored securely, but retrievably, awaiting development of new reactor designs and a political consensus more favorable to reprocessing or a decision to implement permanent sequestration. Whichever path is followed, when spent fuel is removed from a reactor, it is submerged in pools of water. This is done because the fuel is very hot, both in temperature and radioactivity. The water serves both to cool the spent fuel and to shield the environment from radiation. After 10 to 20 years, the shortest-lived and most active components decay away, leaving the spent fuel cooler and less radioactive.
