FAS | Nuke | Intro | Nuke || Search | Join FAS

Plutonium Production

Plutonium was the second transuranium element of the actinide series to be discovered. The isotope 238Pu was produced in 1940 by Seaborg, McMillan, Kennedy, and Wahl by deuteron bombardment of uranium in the 60-inch cyclotron at Berkeley, California. Plutonium also exists in trace quantities in naturally occurring uranium ores. It is formed in much the same manner as neptunium, by irradiation of natural uranium with the neutrons which are present.

Plutonium has assumed the position of dominant importance among the trasuranium elements because of its successful use as an explosive ingredient in nuclear weapons and the place which it holds as a key material in the development of industrial use of nuclear power. One kilogram is equivalent to about 22 million kilowatt hours of heat energy. The complete detonation of a kilogram of plutonium produces an explosion equal to about 20,000 tons of chemical explosive. Its importance depends on the nuclear property of being readily fissionable with neutrons and its availability in quantity. The world's nuclear-power reactors are now producing about 20,000 kg of plutonium/yr. By 1982 it was estimated that about 300,000 kg had accumulated. The various nuclear applications of plutonium are well known. 238Pu has been used in the Apollo lunar missions to power seismic and other equipment on the lunar surface. As with neptunium and uranium, plutonium metal can be prepared by reduction of the trifluoride with alkaline-earth metals.

The metal has a silvery appearance and takes on a yellow tarnish when slightly oxidized. It is chemically reactive. A relatively large piece of plutonium is warm to the touch because of the energy given off in alpha decay. Larger pieces will produce enough heat to boil water. The metal readily dissolves in concentrated hydrochloric acid, hydroiodic acid, or perchloric acid. The metal exhibits six allotropic modifications having various crystalline structures. The densities of these vary from 16.00 to 19.86 g/cm3.

Because of the high rate of emission of alpha particles and the element being specifically absorbed on bone the surface and collected in the liver, plutonium, as well as all of the other transuranium elements except neptunium, are radiological poisons and must be handled with very special equipment and precautions. Plutonium is a very dangerous radiological hazard. Precautions must also be taken to prevent the unintentional formulation of a critical mass. Plutonium in liquid solution is more likely to become critical than solid plutonium. The shape of the mass must also be considered where criticality is concerned.

Uranium and plutonium are composed of several isotopes, some of which are fissile. To produce an explosive device for military purposes requires the percentage of fissile isotopes (U-235 for uranium, Pu-239 for plutonium) present in the material to be of the order of 93%. The levels reached in the nuclear power industry are, however, much lower; less than 5% for uranium and between 50 and 60% for plutonium.

Plutonium containing high quantities of fissile material i.e. Pu-239 in the order of 90-95 %, is known as weapon-grade plutonium. Plutonium containing lower concentrations, in the range of 50-60 % is known as reactor-grade plutonium. The definitions of the various plutonium grades are expressed as a percentage of the isotope Pu-240 which is considered as an impurity for weapons manufacturers.

Plutonium-grade % of Plutonium- 240
Super-grade < 3%
Weapon-grade < 7%
Fuel-grade between 7% and 18% incl.
Reactor-grade > 18%

Reactor-grade plutonium is produced in the core of a reactor when uranium-238 is irradiated with neutrons. Unlike weapon grade plutonium (which is relatively pure plutonium-239), reactor grade plutonium is a mixture of plutonium-238, 239, 240, 241 and 242. It is this mixture of isotopes which renders reactor grade plutonium less unsuitable as a weapon-grade material.

The even numbered isotopes (plutonium-238, 240 and 242) fission spontaneously producing high energy neutrons and a lot of heat. In fact, the neutron and gamma dose from this material is significant and the heat generated in this way would melt the high-explosive material needed to compress the critical mass prior to initiation. The neutrons can also initiate a premature chain reaction thus reducing the explosive yield, typically to a few percent of the nominal yield, sometimes called the "fizzle yield". Such physical characteristics make reactor-grade plutonium extremely difficult to manipulate and control and therefore explain its unsuitability as a bomb-making ingredient.

The odd numbered isotope, plutonium-241, is also a highly undesirable isotope as it decays to americium-241 which is an intense emitter of alpha particles, X and gamma rays. Plutonium-241 has a half-life of 13.2 years which means americium-241 accumulates quickly causing serious handling problems.

Weapon-grade plutonium has different characteristics. It contains mainly Pu-239 which has a half-life of 24 000 years and only very small quantities of Pu-241 (unlike reactor-grade plutonium which can contain around 15% Pu-241.) It is thus relatively stable and can be safely handled with a pair of thick gloves

To achieve the high percentages of Pu-239 required for weapon grade plutonium, it must be produced specifically for this purpose. The uranium must spend only several weeks in the reactor core and then be removed. For this to be carried out in a LWR - the prevalent reactor design for electricity generation - the reactor would have to be shut down completely for such an operation; this is easily detectable.

The Isotopic Composition of Reactor and Weapon Grade Plutonium

  Pu-238 (%) Pu-239 (%) Pu-240 (%) Pu-241 (%) Pu-242 (%)
Reactor-Grade Plutonium
(3,7% U-235, 43,000 MWd/t)1
2 53 24 15 6
Weapon-Grade Plutonium   93 7    

1 Source: Plutonium Fuel - OECD Report, 1989.

Plutonium, one of the two fissile elements used to fuel nuclear explosives, is not found in significant quantities in nature. Plutonium can only be made in sufficient quantities in a nuclear reactor. It must be "bred," or produced, one atomic nucleus at a time by bombarding 238 U with neutrons to produce the isotope 239 U, which beta decays (half-life 23 minutes), emitting an electron to become the (almost equally) radioactive 239 Np (neptunium). The neptunium isotope again beta decays (half-life 56 hours) to 239 Pu, the desired fissile material. The only proven and practical source for the large quantities of neutrons needed to make plutonium at a reasonable speed is a nuclear reactor in which a controlled but self-sustaining 235 U fission chain reaction takes place. The graphite-moderated, air- or gas-cooled reactor using natural uranium as its fuel was first built in 1942. Scale-up of these types of reactors from low power to quite high power is straightforward. ccelerator-based transmutation to produce plutonium is theoretically possible, and experiments to develop its potential have been started, but the feasibility of large-scale production by the process has not been demonstrated.

The "size" of a nuclear reactor is generally indicated by its power output. Reactors to generate electricity are rated in terms of the electrical generating capacity, MW(e), meaning megawatts of electricity. A more important rating with regard to production of nuclear explosive material is MW(t), the thermal power produced by the reactor. As a general rule, the thermal output of a power reactor is three times the electrical capacity. That is, a 1,000 MW(e) reactor produces about 3,000 MW(t), reflecting the inefficiencies in converting heat energy to electricity.

A useful rule of thumb for gauging the proliferation potential of any given reactor is that 1 megawatt-day (thermal energy release, not electricity output) of operation produces 1 gram of plutonium in any reactor using 20-percent or lower enriched uranium; consequently, a 100 MW(t) reactor produces 100 grams of plutonium per day and could produce roughly enough plutonium for one weapon every 2 months. Light-water power reactors make fewer plutonium nuclei per uranium fission than graphite-moderated production reactors.

In addition to production of plutonium, nuclear reactors can also be used to make tritium, 3 H, the heaviest isotope of hydrogen. Tritium is an essential component of boosted fission weapons and multi-stage thermonuclear weapons. The same reactor design features which promote plutonium production are also consistent with efficient tritium production, which adds to the proliferation risk associated with nuclear reactors.

Reactors are generally purpose-built, and reactors built and operated for plutonium production are less efficient for electricity production than standard nuclear electric power plants because of the low burnup restriction for production of weapons grade plutonium. The types nuclear fission reactors which have been found most suitable for producing plutonium are graphite-moderated nuclear reactors using gas or water cooling at atmospheric pressure and with the capability of having fuel elements exchanged while on line. Several distinct classes of reactor exist, each optimized for one purpose, generally using fuel carefully chosen for the job at hand. These classes include the following:

The first nuclear reactor, CP-1, went critical for the first time on 2 December 1942 in a squash court under Stagg Field at the University of Chicago. Construction on CP-1 began less than a month before criticality was achieved; the reactor used lumped uranium metal fuel elements moderated by high-purity graphite. Within 2 years the United States first scaled up reactor technology from this essentially zero-power test bed to the 3.5 MW (thermal) X-10 reactor built at Oak Ridge, Tennessee, and then again to the 250-megawatt production reactors at Hanford. The Hanford reactors supplied the plutonium for the Trinity test and the Nagasaki war drop. Clearly, reactor technology does not stress the capabilities of a reasonably well-industrialized state at the end of the twentieth century.

Some problems did arise with the scale-up to hundreds of megawatts: the graphite lattice changed crystal state, which caused some deformation, and the buildup of a neutron-absorbing xenon isotope poisoned the fission reaction. This latter problem was curable because of the foresight of the duPont engineers, who built the reactor with many additional fuel channels which, when loaded, increased the reactivity enough to offset the neutron absorption by the xenon fission product.

Finally, the problem of spontaneous emission of neutrons by 240 Pu produced in reactor plutonium became apparent as soon as the first samples of Hanford output were supplied to Los Alamos. The high risk of nuclear pre-initiation associated with 240 Pu caused the abandonment of the notion of a gun-assembled plutonium weapon and led directly to the adoption of an implosion design.

Since each fission produces only slightly more than two neutrons, on average, the neutron "economy" must be managed carefully, which requires good instrumentation and an understanding of reactor physics, to have enough neutrons to irradiate useful quantities of U-238. Note, however, that during the Manhattan Project the United States was able to scale an operating 250 watt reactor to a 250 megawatt production reactor. Although the instrumentation of the day was far less sophisticated than that in use today, the scientists working the problem were exceptional. A typical production reactor produces about 0.8 atoms of plutonium for each nucleus of U-235 which fissions.

A typical form of production reactor fuel is natural uranium metal encased in a simple steel or aluminum cladding. Because uranium metal is not as dimensionally stable when irradiated as is uranium oxide used in high burnup fuel, reactors fueled with the uranium metal must be confined to very low burnup operation, which is not economical for electricity production. This operational restriction for uranium metal fuel results in the production of plutonium with only a small admixture of the undesirable isotope, 240 Pu. Thus, it is almost certain that a reactor using metallic fuel is intended to produce weapons grade plutonium, and operation of such a reactor is a strong indicator that proliferation is occurring.

A Heavy Water Reactor would be based on a low-pressure, low-temperature application of nuclear fission technology specifically designed to produce plutonium [or tritium]. The reactor vessel and cooling system configuration (with primary and secondary cooling loops) would be similar to that used in commercial light water reactor nuclear power technology. The HWR would use heavy water as the reactor coolant and moderator. Heavy water, circulated through the core for cooling and moderation, also passes through heat exchangers that are external to the reactor tank. The heat is in turn carried away by the secondary cooling system. The heavy water in the tank surrounding the fuel would represent the bulk moderator.

The cooling system transferring the heat from the reactor to the heat sink can be configured in a wide variety of designs ranging from fresh-water cooling, through evaporative systems, to dry cooling, including some of their combinations. Reactors are either cooled by using saltwater from coastal (estuarine), freshwater cooling pond, lake water or wet cooling towers, depending on the availability of water resources at the particular site.
1 watt = 3.412 BTU per hr
1 MWt = 3,412,000 BTU per hr
1 Ton (heat load)   = 12,000 BTU per hr
1 Cooling Tower Ton = 15,000 BTU per hr
  BTU = GPM of Water x 500 x Temperature Difference [degrees F]
1 BTU = .293 watt - hour 

The heat dissipation system selected, wet or dry, would be dependent on site characteristics. Both wet and dry cooling systems would use water as the heat exchange medium. Wet systems would use water towers and the evaporation process to carry off heat. Dry systems, designed for cold climates, would use water in closed nonevaporative cooling towers to carry off heat to the atmosphere by conduction through radiator-like vanes. In moderate climates, fans would be added to the dry cooling towers to move air over the vanes. There would be some water loss through evaporation in a dry system, but significantly less than with a wet tower. Dry cooling towers would be used for the reactors at all dry sites.

For both once-through and closed-cycle cooling systems, the water intake and discharge structures are of various configurations to accommodate the source water body and to minimize impact to the aquatic ecosystem. The intake structures are generally located along the shoreline of the body of water and are equipped with fish protection devices. The discharge structures are generally of the jet or diffuser outfall type and are designed to promote rapid mixing of the effluent stream with the receiving body of water. Biocides and other chemicals used for corrosion control and for other water treatment purposes are mixed with the condenser cooling water and discharged from the system.

With lake the cooling water from the lake is pumped through a large number of heat exchanger tubes. The cooling water is heated during this process, and is then returned to the lake. The predominant water use at a nuclear reactor is for removing excess heat generated in the reactor. The quantity of water used is a function of several factors, including the capacity rating of the plant and the increase in cooling water temperature from the intake to the discharge. The larger the plant, the greater the quantity of waste heat to be dissipated, and the greater the quantity of cooling water required.

In a once-through cooling system, circulating water for condenser cooling is drawn from an adjacent body of water, such as a lake or river, passed through the condenser tubes, and returned at a higher temperature to the adjacent body of water. The waste heat is dissipated to the atmosphere mainly by evaporation from the water body and, to a much smaller extent, by conduction, convection, and thermal radiation loss.

Recirculating cooling systems consist of either natural draft or mechanical draft cooling towers, cooling ponds, cooling lakes, or cooling canals. Because the predominant cooling mechanism associated with closed-cycle systems is evaporation, most of the water used for cooling is consumed and is not returned to a water source.

In closed-cycle systems, the cooling water is recirculated after the waste heat is removed by dissipation to the atmosphere, usually by circulating the water through large cooling towers constructed for that purpose. Cooling towers are needed when a body of water large enough to provide the cooling, or groundwater is not available in sufficient quantity and there are no other suitable surface water sources available. Closed-cycle cooling towers represents a type of cooling tower that includes both dry cooling towers and hybrid wet/dry cooling towers. Increased cooling tower performance can be achieved by adding surface area or by boosting the flow rate. The former is considerably more expensive than the latter since flow rate can be increased by employing a bigger fan motor allowing increased fan speed.

Wet cooling towers use the same condenser system as in lake cooling, however, the cooling water comes from a large basin at the bottom of the cooling tower. Wet cooling towers use freshwater and achieve 80% of their cooling by evaporation of the cooling water. This evaporation represents a loss of millions of liters of water per year, and dry cooling may be a more attractive option for cooling.

In a natural draft cooling tower the heated cooling water from the condenser is sprayed down the inside of the cooling tower whilst air, under the effects of natural convection flows up through the cooling tower. The air draft evaporates some of the cooling water, lowering the temperature of the remaining cooling water. The draft can also be produced by fans in an induced draft cooling tower. A plume of pure water vapor can often be seen exiting the top of wet cooling towers particularly when the atmosphere has high humidity.

A mechanical draft cooling tower cools circulating water. In the cooling tower, circulating water is directed to the top of the tower and then flows downward through the tower while induced draft fans draw ambient air upward. Heat is transferred to the ambient air primarily through the evaporation of a significant portion of the cooling water. At certain times during the year, a visible plume rises from the tower due to this evaporation cooling process.

A dry cooling tower uses significantly less water. There are two main types of dry cooling technology, the direct system and the indirect system. An air-cooled system operates like a very large automobile radiator. These systems use a flow of air to cool water flowing inside finned tubes. It is essentially a closed loop system where air is passed over large heat exchange surfaces. While air cooling is a reliable and proven technology, it has some technical and economic drawbacks in comparison to a wet mechanical cooling system, which requires the use of significant amounts of water. The principal drawbacks of air cooling are increased noise levels, higher capital costs and larger physical dimensions.

Indirect dry cooling towers use the same condenser system as in lake cooling, however the cooling water is recirculated through banks of finned tubes over which cooling air flows. The air flow is induced in a natural draft cooling tower by convection. The natural draft cooling tower for dry cooling is larger than for an equivalent wet system, since heat transfer rates are much less.

The direct dry cooling mechanical draft cooling tower consists of a concrete structure supporting the mechanical draft fans and exhaust plenum. If fans are used instead of the natural draft tower, a large number of fans would be required to achieve the same heat rejection. This is because the temperature difference between the air and the cooling water is relatively small.

Indirect dry cooling systems tend to have a large capital cost (due to the large cost of the natural draft cooling tower) but low operating cost. In comparison direct dry cooling has a low capital cost but high operating cost (due to large power consumption of the fans). In general, direct dry cooling is favored at sites with low fuel cost (the fan power is less costly), while indirect dry cooling is more suitable at sites with high fuel costs.

The cooling system needs to be located as near as possible to the reactor. It is possible to locate a fan forced cooling system closer to the reactor than a natural draft system. The tube banks may be located directly adjacent to the reactor hall, minimising piping distance, and the fans are located under the tubebanks.

The sources of routine radioactive gaseous emissions to the atmosphere are the air ejector which removes noncondensable gases from the coolant, and gaseous and vapor leakages, which, after monitoring and filtering, are discharged to the atmosphere via the building ventilation systems. The off-gas treatment system collects noncondensable gases and vapors that are exhausted at the condenser via the air ejectors. These off-gases are processed through a series of delay systems and filters to remove airborne radioactive particulates and halogens, thereby minimizing the quantities of the radionuclides that might be released. Building ventilation system exhausts are another source of gaseous radioactive wastes.


Unlike fuel from fossil plants that discharge ash with negligible heat content, fuel discharged from nuclear reactors contains appreciable quantities of fissile uranium and plutonium ("unburned" fuel). These fuel elements must be removed from a reactor before the fissile material has been completely consumed, primarily because of fission product buildup. Fission products capture large numbers of neutrons, which are necessary to sustain a chain fission reaction. In the interest of economic utilization of nuclear fuels and the conservation of valuable resources, several countries have constructed reprocessing plants to recover the residual uranium and plutonium values, utilizing a variety of physical and chemical methods.

Spent fuel contains fission products and actinides produced when nuclear fuel is irradiated in reactors, as well as any unburned, unfissioned nuclear fuel remaining after the fuel rods have been removed from the reactor core. After spent fuel is removed from reactors, it is stored in racks placed in storage pools to isolate it from the environment. An assumption of a cooling time of 160 days between the discharge of spent fuel from the reactor and the reprocessing of the fuel is based on the optimum for recycling plutonium as well as uranium.

Plutonium is removed from spent fuel by chemical separation; no nuclear or physical separation (as for example in uranium enrichment) is needed. To be used in a nuclear weapon, plutonium must be separated from the much larger mass of non-fissile material in the irradiated fuel.

After being separated chemically from the irradiated fuel and reduced to metal, the plutonium is immediately ready for use in a nuclear explosive device. If the reactor involved uses thorium fuel, 233 U, also a fissile isotope, is produced and can be recovered in a process similar to plutonium extraction.

The first plutonium extraction (reprocessing) plants to operate on an industrial scale were built at Hanford, Washington, during the Manhattan Project. The initial plant was built before the final parameters of the extraction process were well defined.

Reprocessing plants are generally characterized by heavy reinforced concrete construction to provide shielding against the intense gamma radiation produced by the decay of short-lived isotopes produced as fission products. Plutonium extraction and uranium reprocessing are generally combined in the same facility in the civilian nuclear fuel cycle. Although the United States no longer reprocesses civil reactor fuel and does not produce plutonium for weapons, other countries have made different choices. Britain, France, Japan, and Russia (among others) operate reprocessing plants.

Heavy industrial construction. All operations are performed in a facility that is usually divided into two structural sections (hardened and nonhardened) and two utility categories (radiation and ventilation/contamination). The hardened portion of the building (reprocessing cells) is designed to withstand the most severe probable natural phenomena without compromising the capability to bring the processes and plant to a safe shutdown condition. Other parts of the building (i.e., offices and shops), while important for normal functions, are not considered essential and are built to less rigorous structural requirements.

Radiation is primarily addressed by using 4- to 6-ft thick, high-den-sity concrete walls to enclose the primary containment area (hot cells). A proliferator who wishes to reprocess fuel covertly for a relatively short time -- less than a year would be typical -- may use concrete slabs for the cell walls. Holes for periscopes could be cast in the slabs. This is particularly feasible if the proliferator cares little about personnel health and safety issues.

Fuel storage and movement. Fuel is transported to the reprocessing plant in specially designed casks. After being checked for contamination, the clean fuel is lowered into a storage pool via a heavy-duty crane. Pools are normally 30-ft deep for radiation protection and contain a transfer pool, approximately 15-ft deep, that provides an underwater system to move the fuel into an adjacent hot cell.

Fuel disassembly. Fuel elements are breached (often chopped) to expose the fuel material for subsequent leaching in nitric acid (HNO 3 ). Fuel cladding is frequently not soluble in nitric acid, so the fuel itself must be opened to chemical attack.

Fuel dissolution. Residual uranium and plutonium values are leached from the fuel with HNO 3 . The cladding material remains intact and is separated as a waste. The dissolver must be designed so that no critical mass of plutonium (and uranium) can accumulate anywhere in its volume, and, of course, it must function in contact with hot nitric acid, a particularly corrosive agent. Dissolvers are typically limited-life components and must be replaced. The first French civilian reprocessing plant at La Hague, near Cherbourg, had serious problems with leakage of the plutonium-containing solutions. Dissolvers may operate in batch mode using a fuel basket or in continuous mode using a rotary dissolver (wheel configuration).

Fissile element separation. The PUREX (Plutonium Uranium Recovery by EXtraction) solvent extraction process separates the uranium and plutonium from the fission products. After adjustment of the acidity, the resultant aqueous solution is equilibrated with an immiscible solution of tri-n-butyl phosphate (TBP) in refined kerosene. The TBP solution preferentially extracts uranium and plutonium nitrates, leaving fission products and other nitrates in the aqueous phase. Then, chemical conditions are adjusted so that the plutonium and uranium are reextracted into a fresh aqueous phase. Normally, two solvent extraction cycles are used for the separation; the first removes the fission products from the uranium and plutonium, while the second provides further decontamination. Uranium and plutonium are separated from one another in a similar second extraction operation. TBP is a common industrial chemical used in plasticizers and paints. Solvent extraction usually takes place in a pulse column, a several-inch diameter metal tube resistant to nitric acid and used to mix together the two immiscible phases (organic phase containing TBP and an aqueous phase containing U, Pu, and the fission products). The mixing is accomplished by forcing one of the phases through the other via a series of pulses with a repetition rate of 30 to 120 cycles/minute and amplitudes of 0.5 to 2.0 inches. The metal tube contains a series of perforated plates which disperses the two immiscible liquids.

U & Pu product purification. Although plutonium and uranium from sol-vent extraction are nearly chemically pure, additional decontamination from each other, fission products, and other impurities may be required. Large plants use additional solvent extraction cycles to provide this service, but small plants may use ion exchange for the final purification step (polishing).

Metal preparation. Plutonium may be precipitated as PuF 3 from aqueous nitrate solution by reducing its charge from +4 to +3 with ascorbic acid and adding hydrofluoric acid (HF). The resulting solid is separated by filtration and dried. Reprocessed uranium is rarely reduced to the metal, but it is converted to the oxide and stored or to the hexafluoride and re-enriched. Plutonium (and uranium) metal may be produced by the reaction of an active metal (calcium or magnesium) with a fluoride salt at elevated temperature in a sealed metal vessel (called a "bomb"). The metal product is freed from the slag, washed in concentrated HNO 3 to remove residue, washed with water, dried, and then remelted in a high temperature furnace (arc).

Waste treatment/recycle. Reprocessing operations generate a myriad of waste streams containing radioactivity. Several of the chemicals (HNO 3 ) and streams (TBP/kerosene mixture) are recycled. All streams must be monitored to protect against accidental discharge of radioactivity into the environment. Gaseous effluents are passed through a series of cleaning and filtering operations before being discharged ,while liquid waste streams are concentrated by evaporation and stored or solidified with concrete. In the ultimate analysis, the only way to safely handle radioactivity is to retain the material until the activity of each nuclide disappears by natural decay.

Early plants used "mixer-settler" facilities in which the two immiscible fluids were mixed by a propeller, and gravity was used to separate the liquids in a separate chamber. Successful separation requires that the operation be conducted many times in sequence. More modern plants use pulse columns with perforated plates along their length. The (heavier) nitric acid solution is fed in at the top and the lighter TBP-kerosene from the bottom. The liquids mix when they are pulsed through the perforations in the plates, effectively making a single reactor vessel serve to carry out a series of operations in the column. Centrifugal contactors using centrifugal force have also been used in place of mixer-settlers. The process must still be repeated many times, but the equipment is compact. New plants are built this way, although the gravity-based mixer-settler technology has been proven to be satisfactory, if expensive and space-consuming.

A single bank of mixer-settler stages about the size of a kitchen refrigerator can separate enough plutonium for a nuclear weapon in 1-2 months. A bank of eight centrifugal contactors can produce enough plutonium for an explosive device within a few days and takes up about the same space as the mixer-settler. Hot cells with thick radiation shielding and leaded glass for direct viewing, along with a glove box with minimal radiation shielding, are adequate for research-scale plutonium extraction, are very low technology items, and would probably suffice for a program designed to produce a small number of weapons each year. The concrete canyons housing many smaller cells with remotely operated machinery are characteristic of large-scale production of plutonium.

When plutonium is produced in a nuclear reactor, inevitably some 240 Pu (as well as heavier plutonium isotopes, including 241 Pu and 242 Pu) is produced along with the more desirable 239 Pu. The heavier isotope is not as readily fissionable, and it also decays by spontaneous fission, producing unwanted background neutrons. Thus, nuclear weapon designers prefer to work with plutonium containing less than 7 percent 240 Pu.

A method for separating plutonium isotopes could be used to remove the heavier isotopes of plutonium (e.g., 240 Pu) from reactor-grade plutonium, thus producing nearly pure 239 Pu. Uranium isotope separation techniques [e.g., atomic vapor laser isotope separation (AVLIS)] might be applied to this task. However, this would require mastery of production reactor and reprocessing technologies (to produce and extract the plutonium) in addition to isotope enrichment technology (to remove the heavier plutonium isotopes). In practice, it is simpler to alter the reactor refueling cycle to reduce the fraction of plutonium which is 240 Pu.

The plutonium must be extracted chemically in a reprocessing plant. Reprocessing is a complicated process involving the handling of highly radioactive materials and must be done by robots or by humans using remote manipulating equipment. At some stages of the process simple glove boxes with lead glass windows suffice. Reprocessing is intrinsically dangerous because of the use of hot acids in which plutonium and intensely radioactive short-lived fission products are dissolved. Some observers have, however, suggested that the safety measures could be relaxed to the extent that the proliferator deems his technicians to be "expendable." Disposal of the high-level waste from reprocessing is difficult. Any reprocessing facility requires large quantities of concrete for shielding and will vent radioactive gases (Iodine-131, for example) to the atmosphere.


An indigenous uranium mining industry might provide early indication of a clandestine uranium or plutonium-based weapon program and is a sure indicator of at least the possibility. For the plutonium path, natural uranium could fuel a graphite- or heavy-water moderated plutonium-production reactor. A sizable research program involving breeder-reactors or the production of heavy water or ultra-pure carbon and graphite products might also be cause for concern, especially if such programs were not easily justifiable on other accounts.

Small research or power reactors with high neutron flux and significant amounts of uranium-238 in their cores can also be used to produce plutonium. However, a 40 to 50 MW(t) undeclared reactor (enough to produce plutonium for at least one bomb per year) should be easily discernible to overhead infrared sensors, at least if it is built above ground and located away from heavy industrial areas (such a location might be chosen for security and safety reasons anyway).

Inspections of safeguarded reactors, especially if carried out at more random intervals, might detect unnecessary placement of uranium-238 in or around the core, augmenting the rate of plutonium production. Similarly, inspections of CANDU-style reactors (a heavy-water-moderated reactor that can be refueled online) or of frequently shut-down LWRs should call attention to very low-bum-up fuel cycles, from which the plutonium produced is predominantly plutonium-239, the isotope best suited for weapons.

In general, the plutonium-production route, which involves reprocessing of spent reactor-fuel to extract plutonium, would be easier to detect than would be a small-scale clandestine uranium-enrichment facility. Plutonium and uranium from spent fuel (as well as enriched uranium from research reactor cores), is reclaimed by chopping up and dissolving the fuel elements in acid, subjecting the solution to solvent-extraction and ion-exchange processes, and chemically converting the plutonium and uranium in the resulting liquids to metallic or oxide forms. Methods for doing this, including the most common one, known as PUREX, involve various well-understood chemical processes that use characteristic groups of materials.

Detection of these materials, either by environmental sampling or by impactions, could indicate reprocessing activity. Some chemicals might also be observed through export monitoring; for example, high-purity calcium and magnesium, which are used in the metal-conversion step, are included in the Nuclear Supplier Group's new list of sensitive dual-use items to be subjected to export controls. In addition to the characteristic chemicals used in the PUREX process, effluents from reprocessing plants will contain telltale radioactive fission products, including radioactive isotopes of the noble gases xenon and krypton -- especially krypton-85 -- and possibly argon. Measurements made at the U.S. reprocessing facility at the Savannah River Plant in South Carolina have suggested that krypton-85 may be detectable, even from small facilities, at ranges of 10 kilometers or more.

Analysis of plutonium samples or effluents from reprocessing could provide further evidence of weapon intent by revealing the fuel's irradiation level. For most types of reactor, a very low fuel-irradiation level would be a strong indicator of weapon activity. In addition, isotopic correlation techniques -- which compare the isotopic ratios of different samples of plutonium -- can provide sensitive indicators of plutonium production history or material diverted from one facility to another.

Sources and Methods

Additional Resources

FAS | Nuke | Intro | Nuke || Search | Join FAS

Originally created by John Pike
Updated Tuesday, June 20, 2000 7:02:38 AM