by Ben Brabson, Indiana University, Nov 18, 2007
Light Water (H2O),
Heavy Water (D2O), Breeder, and Fast
Neutrons hitting one of these nuclei creates a fission and more neutrons. Fissile fuels, 23592U, 23994Pu, 23392U fission when hit by thermal neutrons. Natural Uranium is 99.3% U-238, 0.7% U-235. A fuel must be enriched to ~3.0% U-235 to sustain a chain reaction with H20 as coolant and moderator. Uranium occurs naturally, while Pu-239 must be bread from the fertile fuel U-238, and U-233 from the fertile fuel Th-232. The energy provided is about 200 MeV/fission.
A coolant, carries heat from the core to the turbine. H20 has high specific heat, is inexpensive, but does absorb some neutrons (sabsorption = 0.332 bn). D20 also has high specific heat, does not absorb many neutrons (sabsorption = 5.2 x 10-4 bn), but is very expensive.** Helium gas can circulate through a very hot core giving higher efficiency and is used in the High Temperature Gas-cooled Reactors (HTGR) but must be highly compressed to increase its specific heat. Sodium is a metal that melts at 98 0C, and has large specific heat, and it absorbs almost no neutrons, but highly corrosive and explosive in contact with water. It makes an excellent coolant for fast reactors and breeder reactors.
A moderator slows neutrons from >1 MeV kinetic energy to ~ 0.025 eV (thermal). Typically, moderators contain small nuclei. The neutron kinetic energy is shared in collisions of the fast neutrons with nuclei of comparable size. H20 is an excellent moderator because of the hydrogen nuclei (protons) in the water. As a moderator, D20 is almost as good as normal water and has the added advantage that its neutron absorption cross section is small. Graphite also slows neutrons well, is inexpensive, but burns (Chernobyl).
Are made from excellent absorbers of neutrons, are used both to control the chain reaction during normal operation and to shut down the nuclear process. As fissile fuel is depleted, the control rods are pulled farther out of the core. Boron, 115B, and Cadmium, 11248Cd, are both excellent absorbers of neutrons. The control rods surround the fuel rods and are inserted from the bottom of the reactor. The neutron absorption cross section of Boron is ~775 bn and of Cadmium ~20,000 bn.
Light Water Reactor Problems/Dangers:
1. Proliferation: The reprocessing of spent fuel rods for the Pu239, leading to proliferation of nuclear weapons.
2. Loss Of Coolant Accidents (LOCA) - Three Mile Island, Chernobyl
3. High Level Waste Disposal - spent fuel rods with highly radioactive fission fragments. [Yucca Mountain, Nevada]
4. Fuel Transport from point of enrichment to reactor (hijack, accident)
5. Low Level Emissions during normal operation - Iodine-131 ...
Sources of Heavy Water (D2O)
One hydrogen atom in ~7000 in natural water has a deuteron nucleus (p+n). Heavy water is distilled from light water by taking advantage of the heavier mass of D2O. For example, the boiling point of heavy water is 101.4 oC compared with 100 oC for normal water. Similarly, the mass difference between deuterium and hydrogen can be used in the separation process. Deuterium(21H) boils at -249.7 oC while hydrogen (11H) boils at -252.5 oC. Small differences in the affinities of deuterium and hydrogen for various compounds can also be exploited (See Girdler Sulfide process, for example). Heavy water allows the use of natural uranium in a fission reactor, thus avoiding the complex process of U235 enrichment. Also, tritium (31H) can be produced from (21H). [Why is this important?] The major producers of D2O are Canada, Argentina, India, and Norway.
Advanced Reactor Designs
Gail H. Marcus and Alan E. Levin (Physics Today, April 2002, pp. 54-60) give an excellent summary of new designs for LWRs. They define the generations of reactor design roughly as follows. Generation I refers to the early prototype reactors in the 1950s. Generation II refers to the comercial power reactors, both BWRs (25% of US), and PWRs (75% of US).
Generation III are light water reactor designs with advanced
safety features that are passive. They will rely on stored energy
and natural processes rather than electric power in case of an
accident. Generation III designs include General Electric's Advanced
BWR (ABWR), Westinghouse's System 80+ and Advanced Passive 600 MWe
(AP600). They attempt to address four problems, a) cost, b) safety, c)
proliferation, and d) waste disposal. The designs attempt
to reduce costs by sticking to smaller modular well-tested
designs. Poliferation refers to the diversion of the produced
Pu-239 to nuclear weapons. Guaranteeing countries without nuclear
weapons a constant supply of nuclear fuel and waste disposal for their
reactors is one strategy. Generation III reactors are also
designed to have longer "burn" cycles (1 year --> 18 months or 2
years) with greater utilization of the uranium and correspondingly less
Generation III+ designs include Westinghouse's AP1000, a larger
version of the AP600, and IRIS (International Reactor Innovative and
Secure), an integral PWR with
the reactor and steam generators in the same vessel. Two
gas-cooled reactor designs are also Generation III+. The German
designed Pebble-bed modular reactors
(PBMR) have "tennis ball sized graphite spheres with poppy seed sized
spheres of Uranium Oxide" cooled by helium gas, not water. General
Atomic has developed a 285 MWe gas turbine modular helium reactor
(GT-MHR) with a "prismatic core" of stacks of hexagonal graphite blocks
with embedded microspheres of uranium, all helium gas cooled. A
similar design was used in the Fort St. Vrain, Colorado high
temperature gas cooled reactor from 1979-1989. Generation III+ designs
are expected in plants built in the next 10 years.
2030 - 2050
Generation IV designs are speculative, using strategies not yet
tested. The designs strive to be simpler, cheaper, more reliable,
and more efficient. They would look to achieve higher operating
temperatures with corresponding higher efficiencies, fewer moving
parts, lower fuel inventory and plutonium buildup. This generation
considers the use of reactors for other purposes including providing
hot water, desalinization of seawater, and hydrogen production for
hydrogen powered vehicles. You might consider taking one of these
topics for your presentation. Ideas include:
a. heat exchange through the reactor wall rather than through the exchange of a coolant
b. collecting the energy of the fission fragments directly, as in an electric cell or magnetic collimator
c. gas cores including uranium vapor vortex flow and a closed magnetohydrodynamic power generation cycle
d. liquid cores including molten salts of thorium or liquid uranium fluorides
e. a fast reactor using sodium evaporation cooling and soidum vapor gas turbines
f. using alkali metal thermal-to-electrical conversion
Breeder reactors are designed to use extra neutrons to convert fertile fuels (U-238, Th-232) into fissile fuels (Pu-239, U-233). The Super-Phoenix in Malville, France is a breeder reactor. In a normal fission process, the number of neutrons produced depends on the incident neutron energy. Fast neutrons (KE ~ 1 MeV) produce typically more than 2 neutrons in each fission. A breeder reactor has no moderator. The neutrons are not slowed down and, of course, the probability of fission is greatly reduced. However, when the fast neutrons manage to produce a fission, more neutrons are produced. Thus, the number of fertile nuclei that can be converted to fissile nuclei is greater if fast neutrons are used.
To compensate for the small fission cross section a number of changes must be made. First, the reactor fuel must be much more highly enriched (~20 to 90 % instead of 3%). Second, the core density is greatly increased. The fuel rods are more closely spaced to increase the probability of hitting a fissile nucleus. Third, the neutron flux (neutrons/m2/sec) is made as high as possible by reducing neutrons absorbed by the coolant. Liquid sodium (T > 300 oC) has a high specific heat and a very low neutron absorption cross section, so is a good choice for coolant. Also, the reduced size of the core helps to reduce the absorption of neutrons by the coolant. "Neutron mirrors" on the edges of the core volume also help to keep the neutron flux high. A neutron mirror is a material like U-235 that produces neutrons when hit by neutrons. Finally, the neutron energy must be adjusted to maximize absorption by the fertile fuels (U-238, Th-232) that b-decay to produce the new fissile fuel. The most effective neutron energy for the conversion of fertile to fissile nuclei is a few keV. Rods of fertile fuel surround the core and are embedded in the core. At the end of a running period (~1 year, for example) the fertile fuel rods are removed and reprocessed to extract the new fissile fuel.
Because of the high density of a breeder, the heat produced is concentrated in a small volume. An advantage is that the core temperature is considerably higher than that of a light water reactor and the corresponding electric generating efficiency correspondingly higher. A disadvantage is that temperature excursions are more difficult to control. The breeder reactor runs closer to the unstable temperature edge than does a light water reactor.
Fast reactors, like breeders, do not have a moderator. The key to their operation is the fact that fast neutrons (~1 MeV) do fission U-235, U-238, Pu-239, etc, albeit with small cross section of order 1 barn. These same fast neutrons do not breed fuel. Only slower (~1 keV) neutrons are effective at breeding fuel. Thus, fast reactors are not designed to produce fuel for another reactor from U-238. On the contrary they are designed to "burn up," that is, to fission the heavy actinides (Z>=89) long-lived waste from other reactors, especially plutonium isotopes and to use U-238 as a fissile fuel. By keeping the neutron energies high (~1 MeV) the amount of breeding of new fuel (Pu-239) is minimized, and U-238 becomes fissile, contributing fission energy. Like the breeders, they must be designed to operate at high density to compensate for the small fission cross section of fast neutrons on any of these actnides. In addition to being able to induce fission of the long-lived actinides, the 1-MeV neutrons in the fast reactor also convert certain long-lived fission fragments to shorter half-lived, more easily stored isotopes. In summary, fast reactors effectively allow the use of U-238 as a fissionable fuel without having to go through the breeding process.
Integral Fast Reactor (IFR) - Argonne National Laboratory near Chicago has a new fast reactor designed to have a highly compact core with enriched fuel. The enriched fuel allows for a longer "burn" period between shutdowns. Compared to a typical burn of ~35,000 MWday/ton for a LWR, the IFR reaches ~100,000 MWday/ton. Thus, a much larger fraction of the U-235 and U-238 are consumed between shutdowns. The IFR is also designed with a negative temperature coefficient for safety. Increased temperature decreases the density of the core. This in turn reduces the neutron flux and the number of fissions. This negative feedback leads to stability, especially in smaller fast reactors. Also, an increase in temperature "Doppler broadens" neutron absorption resonances making them more effective at absorbing neutrons. This also leads to greater stability.
The Plutonium Problem: In the process of slowing neutrons from MeV to thermal energies (0.25 eV), neutrons in the keV range are absorbed by U-238 and are therefore efficient producers of Pu-239 in a normal thermal reactor. Waste from these thermal reactors contains substantial quantities of Pu-239 with 24,000 year half-life and requires either sequestration for 250,000 years, or reprocessing into fuel for a fast reactor. Reprocessing spent fuel from thermal reactors can be done in a couple of ways. In the PUREX process pure Pu is extracted from the mix of plutonium and uranium in waste. This leads to a commerce in pure Pu-239, an extremely dangerous undertaking as Pu-239 is the material of choice for nuclear weapons. A second process called pyrometallurgical processing (Hannum, March & Stanford, Scientific American, December 25, 2005) is much more attractive. Here, the oxides of U and Pu are reduced to metallic form and at high temperature (hence pyro) electroplated onto an electrode. The electroplated metal containing a mix of actinide metals is then reformed into fuel rods for a fast reactor. Essentially all of these actinide metals fission when hit by fast neutrons, including the major constituent, U-238, and the produced Pu-239. Most of the short half-lived fission fragments are removed in the process. The resulting metal fuel rods are unusable for nuclear weapons but well suited to fast reactors. Metal fuel rods run cooler in a reactor than oxide fuel rods, reducing the probability of a meltdown if reactor control is lost. A second attractive feature of this process is that spent metal fuel rods from fast reactors are easily reformed using the same pyrometallurgical process.