by Bob Williams
Since the signing of the Strategic Arms Reduction Treaty (START), the United States and Russia have been forced to confront the novel problem of how to dispose of the fissile weapons materials which both countries previously produced under conditions of maximum effort and urgency. There was no plan in place for the ultimate disposition of this material, so now both countries struggle with the political, technical, and economic consequences of their agreement to reduce nuclear warhead inventories to a small fraction of their prior peaks.
On December 28, 1942, Roosevelt authorized the construction of the first U.S. production(1) plants. Plutonium was produced and recovered at Hanford B, D, and F reactors (all water-cooled, graphite pile designs). Uranium production (enrichment) was started at the Clinton Engineer Works, which was renamed Oak Ridge after the war. Parallel separation(2) efforts were made with gaseous diffusion, Calutrons, and liquid thermal diffusion.
Under the Manhattan project, Hanford reactors produced enough plutonium for the Trinity Site test at Alamogordo, New Mexico and for the Fat Man bomb dropped on Japan. At this same time, diffusion separators in Tennessee yielded enough highly enriched uranium for the Little Boy bomb. Postwar production was relatively low until the construction of new facilities in the 1947-52 time frame. These new facilities were built in Paducah (Kentucky), Oak Ridge (Tennessee), Portsmouth (Ohio), Hanford (Washington), and Savannah River (South Carolina). Processing facilities were built at Fernald (Ohio), Rocky Flats (Colorado), and Pantex (Texas).
Eventually, 16 major facilities and numerous smaller ones were in operation at sites throughout the United States.
These facilities grew to occupy 2.3 million acres of land and 120 million square feed of building floor space.
Production remained high throughout much of the cold war. One of the most critical production requirements was
for tritium(3). By late 1988, the Department of Energy (DOE) established an "urgent
schedule" to build two new tritium production reactors at an estimated cost of $6.8 billion. At about the
same time, serious safety concerns arose at Rocky Flats, Savannah River, and other sites. Rocky Flats was seized
by the FBI, on the basis of unsafe operations, and shut down without any attempt to complete material in progress
or to flush out lines containing chemicals and radioactive isotopes. Savannah River reactors were shut down(4)
for safety reviews.
The events of 1990-91 radically changed the cold war perspective. With the Soviet Union out of existence, the DOE
asked Congress to cancel a pending special isotope separation plant because weapons requirements could be met with
existing inventories. Then the big change: on July 31, 1991 the Strategic Arms Reduction Treaty was signed. START
directed a reduction in warhead stockpiles to 3,500 each for Russia and the United States(5).
The New Production Reactor was delayed by two years, downsized, and eventually canceled. By the end of 1992, DOE
no longer had the critical facilities to produce nuclear weapons(6).
There are now tens of thousands of nuclear weapons to be dismantled and removed from military arsenals. During
the massive cold war buildup, little thought was given to the eventuality that someday the process of massive warhead
building would be reversed. As warheads are disassembled(7), a new inventory of
surplus fissile materials (specifically, highly enriched uranium(8) and plutonium)
is growing. In addition to sources from dismantled weapons, there are contributions to this surplus inventory from
production reactor fuel (fresh, irradiated, and spent), material that was in various stages of manufacture, residues,
and feed stocks. The quantities of surplus weapons material is subject to variation as disposition agreements are
reached, treaties are updated, and vary greatly according to the definitions used. In 1996, DOE declassified most
of the information pertaining to its surplus plutonium and HEU. For purposes of this discussion, it is sufficient
to say that there are tens of metric tons (MT(9)) of plutonium that has been removed
from the weapons program and there are well over 100 MT of HEU.
The very existence of surplus fissile material is a curious situation. For over 50 years, the United States spent
billions of dollars and poured enormous resources into the production of material, which may now have reversed
its value, going from precious to a liability. This is especially true for plutonium. Weapons grade plutonium was
made by an inefficient process (short exposure target irradiation) in the Hanford and Savannah River reactors.
Then, suddenly, the stroke of a pen caused the plutonium inventory to become the equivalent of a huge stockpile
of anthrax. Even more ironic, the negative value of plutonium is largely a feature of U.S. policy; the same material
in Japan or France would presumably have a high positive value and would be used as reactor fuel.
Three options are available for the disposition of surplus fissile materials:
| STORE | convert to stable chemical form | hold in vaults and protect |
| WASTE | stabilize and make unusable for weapons | dispose |
| BURN(10) | convert to reactor fuel | burn in a reactor treat spent fuel as usual |
Storage seems to be a simple and easy solution, but of the three options, it is the first one to be eliminated.
The two obstacle are cost and politics. Cost is high for storage because any material that is weapons usable must
be protected by the highest level of safeguards and security. In order to store plutonium and uranium, existing
metals would have to be converted to the most stabile oxide forms (for example U3O8), then
placed in drums, stored in a nuclear materials vault, and arranged in a safe configuration (to prevent accidental
criticality). The vault would require perpetual protection, inspection, and maintenance. The degree of protection
given to weapons materials is so extreme that the costs are very substantial.
The political objection to storage as a disposition option is that the material could re-enter the weapons inventory
at some future date. The intent of START was to remove the material from the weapons stockpile forever. Even if
storage costs were relatively low, the political pressure to find a one-way path out of the weapons stockpile is
overwhelming.
Plutonium was made by man, at great expense and effort. Getting rid of it may be as difficult. Unlike uranium,
plutonium can be fabricated into a workable bomb even if it contains relatively large amounts of isotopes other
than Pu-239 (the filet mignon). If the material is weapons grade, it takes only a few pounds to make a bomb(11).
To complicate matters further, Pu can be chemically separated from any material that could serve as an appropriate
dilutant. In normal production, Pu is separated from U by a process known as PUREX, which is described in the open
literature. Before Pu can be wasted, it must be rendered completely inaccessible, or highly resistant to a determined
recovery effort (proliferation resistant).
There are two disposal options which have technical and economic merit. The first is to mix the Pu with high level
waste(12) and to then vitrify it into large glass logs, which would be placed in
a geologic repository. This option is comparatively cheap (only a billion taxpayer dollars) and could be accomplished
at a single high level waste disposal site.
The second is to put the material into the bottom 2 kilometers of 4 kilometer deep boreholes. A primary detraction
of the deep borehole option is that the engineering, physics, and chemistry considerations are very complex.
The fun part of plutonium disposal options involves more "creative" approaches. For example, serious
studies have been conducted to evaluate putting the material in sub-seabed mud, underground explosion, co-disposal
with spent reactor fuel, ocean dilution (Ack!), and space launch disposal. These options range from reasonable
and cost efficient to dangerous and unacceptable. The seabed option would involve mixing the Pu with high level
waste and "planting" it in the mud with a penetrator (rocket shaped package, designed to penetrate into
the mud) or by drilling. The Russians have suggested the underground blast. This would not be a warhead-at-a-time
series of blasts, but would involve hundreds or even thousands of pits(13) arranged
around a nuclear device to be vaporized along with tons of surrounding rock (instant glass). The space launch option
is rather obviously impractical, but the plans that were evaluated included placing the plutonium in orbit around
the sun, or causing the material to be drawn into the sun.
MOX Option
In the 1960s the U.S. was planning a nuclear energy program in which fast breeder reactors would produce electrical
power and plutonium; so much plutonium would be produced that the reactors could operate from their own "breeding"
and have excess plutonium which would be burned in light water reactors. This plan was terminated by the Carter
administration, thereby eliminating plutonium as a valuable electricity producing energy source. The reason for
the cancellation was to reduce the possibility of nuclear proliferation. Russia, France and Japan still have plans
to use a plutonium fuel cycle. Meanwhile, other sources of nuclear materials pose orders of magnitude greater probabilities
of undesirable transfer to outlaw nations and terrorists.
Plutonium can be burned in light water reactors, heavy water reactors, or in other types of reactors, such as sodium
cooled fast breeders. Reactor fuel made from plutonium is known as mixed-oxide, or MOX. It consists of depleted(14) or natural uranium mixed with enough plutonium to enable the fuel to function
similarly to conventional commercial uranium fuel. Although there are a couple of reactors in the U.S. which were
designed to use MOX (others can use it, with some limitations), neither the U.S. nor Russia has a MOX fuel fabrication
plant.
The DOE believes that the MOX option is the most attractive alternative for consuming existing and future inventories
of surplus plutonium. There are basically two options for a MOX program. The first approach would use the optimum
amount of plutonium to produce electrical energy (in simple terms, plutonium is substituted for U-235 and used
as commercial fuel). This approach would recover the maximum usable energy from the plutonium, but would take a
comparatively long time to consume the surplus. The second option is to overload the reactor fuel with plutonium
and burn only part of it. This would leave a rather large amount of unburned plutonium in the spent fuel, but would
achieve the goal of making that plutonium highly proliferation resistant (due to the high radiation associated
with spend fuel). This option seems to be in favor because it would effectively destroy the plutonium in the shortest
time. A variation of the second option would be to build a dedicated plutonium burning reactor with no electrical
power production features (dump the heat). This option is unattractive for a couple of reasons, but the overriding
one is that a new reactor would have to be built (reactor prices are expressed in gigabucks).
HEU poses the same proliferation threat as plutonium, but unlike plutonium, it can be converted to a form resembling
its natural source by blending with depleted uranium. If it is blended down to an enrichment of less than 1%, it
can be disposed of as a low-level waste. Unlike plutonium, uranium can easily be used in commercial reactors by
blending it to an enrichment of 4-5% instead of less than 1%. This possibility makes the wasting of uranium (a
bore hole disposal, for example) relatively unattractive. Unfortunately, surplus uranium inventories are composed
of cats and dogs from production reactor fuel, spent fuel, production targets, intermediate process materials,
and HEU that has been removed from warheads. The value of any lot of uranium depends on its assay. If it contains
relatively low concentrations of undesirable isotopes, such as U-234 and U-236, it is considered to be "sweet"
and is potentially valuable. If there are high concentrations of "bad actor" isotopes (these cause high
radiation and high neutron absorption), the economic value of the material may be low enough that it must be disposed
of as waste. In this case, the waste would have to be blended to low level or disposed of by means which are more
appropriate for plutonium.
Fuel Option
The government could recover millions of dollars by selling its inventories of HEU for use as commercial reactor
fuel. As with everything in the nuclear world, this isn't simple. From a practical standpoint, the uranium will
probably have to be blended from HEU to LEU before it could be transferred to a commercial user for conversion
into fuel. Commercial producers of boiling water reactor fuel have learned to blend uranium to desired enrichments
by a process known as jet milling. This highly efficient process involves the mixture of fine particles of uranium
oxide. DOE has rejected this method as a blending option because of fear that someone could figure out how to extract
the U-235 on the basis of particle size or some unknown factor (quite unlikely). That means that the HEU must be
blended in molten, liquid, or gaseous form(15). Of these, liquid blending is the
most likely route(16), but the resulting product would have to be converted to
uranium oxide before it could be used in commercial reactors. There is no facility in the U.S. capable of performing
this conversion, so there are construction costs and red tape obstacles.
Since some of the HEU that is surplus to the weapons program contains the undesirable isotopes (they got there
from fuel reprocessing), it cannot be used in U.S. reactors because it doesn't meet commercial fuel specifications.
This uranium may be burned in European reactors (different specs) or in U.S. reactors where conditions allow management
to accept off-spec fuel. DOE may offer this category of material on a shared savings basis; the utility would calculate
its savings from getting the uranium free and rebate the government for half of this amount.
Compared to the time and cost of dealing with high level nuclear waste, the safe disposal of weapons materials
is going to be cheap and fast. Compared to just about anything else, it is going to be slow and expensive. Uranium
and plutonium were costly and tedious to produce and will be costly and politically frustrating to remove from
our nuclear arsenals.
1. "Production" has a special meaning in nuclear terminology. It refers to facilities
which produce special isotopes. In the case of weapons materials, production is normally accomplished by target
irradiation in nuclear reactors, and by separation techniques for the enrichment of mined uranium.
2. Natural uranium is mostly U-238, but about 0.71 percent of it is U-235. To make nuclear weapons from uranium, it must be greatly enriched in U-235 content. The assays of U.S. weapons uranium are classified, but it is well known that much of the highly enriched uranium used in weapons systems is above 90% U-235.
3. Tritium is an isotope of hydrogen which has a nucleus composed of one proton and two neutrons. It has a relatively short half-life, which causes about 5% of the material to decay every year. Because of this, tritium used in weapons must be periodically removed, purified, and replaced.
4. The five reactors at Savannah River were different from the Hanford piles. The Savannah River reactors were moderated and cooled with heavy water, and were of a totally different configuration. Following the safety shutdown, four of the five reactors were never restarted. K-reactor was singled out for refurbishment and was to produce tritium until the New Production Reactors could be put into service. After the expenditure of about $1 billion, K-reactor was brought to partial power for tests, then shut down permanently.
5. Open literature estimates of U.S. and Russian warhead inventories vary quite a bit. For example, Russia was estimated to have from a few thousand to well over 30,000 warheads. Most estimates credit both countries with stockpiles in the five-figure range.
6. Large inventories of weapons were in service and could be maintained at levels allowed by the treaty; however, the U.S. had no operational facilities to produce new plutonium or tritium, nor could it produce major portions of the physics package in large numbers (because Rocky Flats was out of service).
7. U.S. warheads are assembled from components which are produced at the various highly dispersed weapons facilities. This final assembly is done at Pantex (from "panhandle" and "Texas"), near Amarillo, Texas. The dismantlement process logically happens at the same location.
8. By definition, highly enriched uranium (HEU) is uranium which is enriched to 20% or more. Uranium enriched beyond natural levels, but less than 20%, is defined as low enriched (LEU) and uranium which has an enrichment of less than U-natural is labeled depleted.
10. "Burn" does not refer to combustion, but to consumption by fission in a nuclear reactor.
11. Anyone having the opportunity to visit the nuclear weapons museum in Albuquerque, may have a chance to see the tiny Davy Crockett, which was designed to be carried on a man's back and field launched. The entire package is a bit smaller than a soccer ball.
12. High level waste is probably the nastiest stuff on Earth. It is the waste from reprocessing spent fuel and targets and consists of highly radioactive fission products, hazardous chemicals, and toxic heavy metals. In practice this horrible material resides in storage tanks at sites, such as Hanford and Savannah River, and takes forms of liquid, solid (salt cake), sludge, and powder.
13. Pits are the plutonium product that was produced at Rocky Flats. While the general shape of these items is well known, the dimensions and other parameters are classified.
14. Depleted uranium is a byproduct of uranium enrichment. Massive amounts of depleted uranium remain in U.S. inventories.
15. Molten blending is known as "casting." Liquid blending is usually taken to mean the use of uranyl nitrate. Gaseous blending is done with uranium hexifluoride.
16. The uranium in question is mostly in pure metal or alloy form. The first step in converting it to another form is to dissolve it in nitric acid. It makes economic sense to do the blending at this stage unless the uranium batch happens to be in another form.