IPFM International Panel on Fissile Materials

[This entry is drawn from Chapter Seven of the 2009 Global Fissile Material Report: "Verified Cutoff of Fissile Material Production for Weapons." The printed version includes endnotes and, in some cases, additional figures. Entries are updated to reflect current data.]

Setting up arrangements to verify a ban on the production of fissile materials for weapons is a part of the nuclear disarmament agenda that hopefully will soon be under negotiation at the UN Conference on Disarmament (CD) in Geneva. On 29 May 2009 the CD agreed to begin negotiations on “a non-discriminatory, multilateral and internationally and effectively verifiable treaty banning the production of fissile material for nuclear weapons or other nuclear explosive devices.” The proposed treaty is often referred to as the Fissile Material Cutoff Treaty (FMCT) and by the IPFM as the FM(C)T. Verification of an FM(C)T was discussed at length in Global Fissile Material Report 2008. This chapter provides an overview and places it in the context of the nuclear disarmament agenda. For details, the reader is referred to 2008 report.

Under a nuclear disarmament regime, the distinction between weapon and non-weapon states would disappear and all fissile material would be under international safeguards.
The question is how large a step in that direction will be taken under an FM(C)T. Specifically, negotiation of an FM(C)T will have to address two fundamental issues:

1. Whether and to what extent a treaty banning any new unsafeguarded production of fissile materials should also subject pre-existing non-weapons stocks of fissile material to international monitoring to verify that they are not converted to weapons use, and

2. How such a treaty should be verified, including the extent to which safeguards obligations in the nuclear weapon states and non-nuclear weapon states will converge.

An incomplete moratorium

Four of the five weapons states that are Parties to the Nonproliferation Treaty: the United States, Russia, the United Kingdom and France, declared in the late 1980s and early 1990s that they had permanently ended production of fissile materials for nuclear weapons. China’s government did not make such a public declaration but has let it be known unofficially since the early 1990s that it has suspended production and will only feel compelled to resume if it feels that the effectiveness and/or survivability of its deterrent is being eroded by a buildup of U.S. missile-defenses and/or long-range precision-guided weapons.

In South Asia, production of fissile materials is accelerating as India builds a “minimum deterrent” of unspecified size and Pakistan races to build up its fissile-material production capacity (Chapter 1). Israel’s policy of “opacity,” i.e., not talking about its nuclear-weapon-related activities, has left unclear whether it is continuing to produce weapon-grade plutonium at its Dimona nuclear complex but, most likely it is, if only as a byproduct of its tritium production. Finally, on 24 September 2008, North Korea announced that it would resume separation of plutonium for weapons and, on 13 June 2009, announced that it was launching a program to enrich uranium for weapons as well.

As the world moves toward complete nuclear disarmament, however, all the nuclear weapon states will have to halt production of fissile material for weapons and accept effective arrangements to verify this.

Verification of a ban on production of fissile material for weapons

Verification of a ban on the production of fissile materials for weapons will require determinations that:

1. Production facilities that have been declared shut down are indeed shut down and remain so;

2. All plutonium separated and HEU produced at declared production facilities after the ban comes into force are placed under IAEA safeguards and remain under safeguards; and

3. There are no undeclared enrichment or reprocessing facilities.

Shutdown production facilities. Under an FM(C)T, countries would either convert production facilities ( reprocessing plants, plutonium-production reactors, and enrichment plants) to safeguarded civilian use or shut them down and decommission them.

Reprocessing plants. In practice, the facilities used to recover weapon-grade plutonium from the low-burnup magnesium or aluminum-clad uranium metal used in production reactors are so different from those used to reprocess the high-burnup zirconium-clad uranium-oxide fuel used in most power reactors that no military reprocessing plant has been converted to civilian use. A few plutonium-production reactors have been operated as dual-purpose reactors, producing electricity as well as weapon-grade plutonium, but operating them for electricity production alone has been uneconomic and all such dual-purpose reactors have been decommissioned or soon will be.

Enrichment plants. In the United States, military gaseous diffusion enrichment plants were converted to civilian use but two out of the three have now been shut down and replacement capacity for the third is under construction. In China, it is believed that the two gaseous diffusion plants used to produce HEU for weapons have been shutdown. LEU for China’s power reactors is produced by centrifuge enrichment plants. In France, the Pierrelatte gaseous enrichment plant that produced France’s HEU is being decommissioned. In Russia, three large centrifuge plants that produced HEU for weapons have been converted to producing low-enriched uranium for nuclear power plants. The UK’s centrifuge enrichment plant that produced some of its HEU has similarly been converted.

Most facilities for producing fissile materials for weapons in the five NPT weapon states are therefore shut down and, in some cases, are in the process of being decommissioned (see Appendix 1A).

The verification challenge at these sites will be minimal. It will only be necessary to confirm that key equipment necessary to the operation of the facility has been disabled or removed. Seals could be applied to assure that spent fuel is not introduced into reprocessing plants or uranium feedstock into enrichment plants and remotely monitored electronic cameras and other sensors could be set up to monitor any activity in key areas of the plants with periodic random unannounced on-site checks to make sure that the seals are intact and monitoring systems are functioning properly. Facilities for which there are no conversion plans should be decommissioned as quickly as possible to make their shutdown irreversible.

Operating reprocessing and enrichment plants. The second element of verifying an FMCT would be to assure that any plutonium, HEU or other fissile material produced in a declared reprocessing plant or enrichment plant after the treaty comes into force for a Party, is placed under IAEA safeguards.

Reprocessing. Some weapon states (China, France, India, Russia and the United Kingdom) and one non-weapon state (Japan) are separating large quantities of weapon-usable plutonium from spent power-reactor fuel for civilian purposes. The original rationale was to provide startup fuel for plutonium-breeder reactors. When those reactors were not commercialized, Belgium, France, Germany and Switzerland began to recycle their separated plutonium in light-water-reactor fuel. Japan and China intend to do the same while India and Russia are still moving ahead with their breeder programs, although at a glacial pace. The United Kingdom is winding down its reprocessing and is beginning to consider options for disposing of approximately 100 tons of separated power-reactor plutonium that it has accumulated (Chapter 6).

Reprocessing and plutonium recycle are not economic, nor are plutonium breeder reactors. Nor do they simplify the problem of spent fuel disposal. Furthermore, the spread of reprocessing has been closely associated with the spread of nuclear-weapons programs. Today, only one non-weapon state, Japan, reprocesses and twelve non-weapon states that in the past sent their spent fuel to France, Russia and the United Kingdom to be reprocessed have not renewed their contracts. For them, reprocessing, simply exchanged the problem of storing and disposing of spent fuel for the equally politically challenging problem of storing and disposing of the solidified high-level reprocessing waste that the reprocessing countries insist on sending back to their foreign customers. Countries that have reprocessing plants have the political advantage that it does provide a single central location to which their nuclear power plants can ship their spent fuel.

Modern civilian reprocessing plants are designed to separate annually 7–17 tons of plutonium—enough to make a thousand nuclear weapons or more. Since plutonium is a directly weapon-usable material, this puts a tremendous burden on safeguards. Even with input and output measurement errors of plutonium from reprocessing and mixed-oxide (uranium-plutonium) fuel fabrication plants as low as one percent, it would be impossible to prove by mass balance checks alone that plutonium, sufficient to make tens of weapons had not been diverted. The IAEA, therefore, adds layers of expensive monitoring, containment and surveillance to increase its confidence that no significant diversions are occurring at Japan’s reprocessing plants, especially at the large, recently completed plant at Rokkasho. This reprocessing plant plus a smaller pilot plant, the only reprocessing facilities in a non-weapon state, account for about 20 percent of the IAEA’s total safeguards budget.

Map

At Rokkasho, the IAEA was able to verify the design of the reprocessing plant and installed independent measuring instrumentation before some areas of the plant were embedded in concrete or became contaminated. For pre-existing plants, the IAEA would not have this luxury. Nevertheless, it should be possible to design safeguards procedures, including the use of short-notice random inspections that would make it difficult to operate the plant improperly and make it possible to detect a diversion of plutonium larger than the measurement errors in the plant plutonium throughput.

It would be better for verification of an FMCT, however, if reprocessing were phased out altogether. This would also have the advantage of allowing attention to be focused on the elimination of the existing large stockpiles of civilian and excess weapons plutonium. Given that civilian spent-fuel reprocessing is neither economic nor necessary to nuclear power for the foreseeable future, such a phase-out does not appear an unreasonable goal (Chapter 8).

Enrichment. Only one country, India, is known to be producing HEU for non-weapon purposes today. India is building a naval reactor that reportedly is fueled with HEU enriched to between 20 and 40 percent uranium-235. Other countries (the United States, Russia and the United Kingdom) are known to use HEU in naval-reactor fuel but their requirements could be satisfied for many decades using excess Cold War weapons HEU. France has already shifted its naval reactors to LEU. HEU is also used as a research reactor fuel but, outside Russia at least, it is being replaced by LEU.

Thus the major challenge in the near term would be to verify that all operating enrichment plants except India’s are indeed not producing HEU. In principle, the enrichment of the uranium in the key collector or “header” pipes in the enrichment plants could be monitored. This may be impractical in Russia’s huge enrichment plants, however, because they have hundreds of thousands of relatively low-capacity centrifuges and complex piping arrangements (see Figure 7.1).

Figure 7.1

A supplementary approach to detect clandestine HEU production in a large enrichment plant would be to look for traces of leaked HEU. The IAEA has used this technique with remarkable effect in Iran and elsewhere. It involves taking “swipes” of surfaces inside a facility and then inspecting the dust picked up by the swipe for particles of uranium. When such particles are identified, they can be bombarded by a beam of atoms that will knock off uranium ions that can be passed through a mass spectrometer to determine the percentages of uranium-235 and uranium-238. Figure 7.2 shows an example of a pair of images of a 0.15-mm (150-micron) square that are formed by ions knocked off uranium particles deposited on the surface of a planchet. The relative brightness of the particles in the images depends upon the percentages of uranium-235 or uranium-238 atoms in a given particle.

Figure 7.2

The complication for the case of Russia’s centrifuge enrichment facilities is that there could be old particles of HEU dating back to when Russia was producing HEU before 1989. These particles would have to be distinguished from possible new particles of HEU. One approach, age dating the particles using the in-built clock associated with the decay of uranium-234 into thorium-230 is discussed in Global Fissile Material Report 2008, Chapter 4.

India is still producing HEU but its enrichment plant is small enough so that its output of HEU could be accurately monitored.

Non-weapon use of fissile materials. Once HEU or plutonium is under safeguards, it must be carefully monitored until, in the case of HEU, it is down-blended to low enriched uranium, and, in the case of plutonium, it is embedded in a radioactive matrix equivalent to the plutonium in spent power reactor fuel. In most cases, effective approaches for doing this have been worked out for NPT safeguards in non-weapon states.

A new safeguards issue for the weapon states, however, will be the fact that many of them have HEU-fueled military reactors. Most of these are naval reactors but Russia, for example, also uses HEU-fueled reactors to produce tritium for its nuclear weapons.

Any new production of HEU for reactor fuel would have to be safeguarded under an FMCT and, depending upon the scope of the FMCT, some pre-existing stocks of HEU also could come under safeguards. The quantity of HEU in military-reactor fuel cycles is substantial. The United States, for example, uses an average of about 2000 kg of weapon-grade uranium annually to fuel the reactors that propel its submarines and aircraft carriers. If converted to first-generation Nagasaki-type implosion weapons at 25 kg per weapon, that would be enough to produce 80 nuclear weapons a year.

The non-weapon use of HEU produced or reserved for naval and tritium-production reactor fuel could be verified by measuring the quantity of HEU produced or withdrawn from stocks to make HEU fuel and then confirming that it was actually put into a reactor. Verification procedures that have been developed for HEU-fueled research reactors might have to be altered if, as appears likely, some of the weapon states will consider the designs of their military reactors and their fuel to be sensitive information. The IPFM has been exploring various technical approaches that could help, but the IAEA and the weapon states would have to work out compromises under which the most sensitive design and operating information would be concealed while still enabling the IAEA to obtain enough information to verify that no significant amount of HEU was being diverted. The best solution, however, would be for the weapon states to switch to LEU-fueled reactors. The international community then would not have to worry about possible diversions of HEU from the naval fuel cycles and the nuclear navies could preserve their privacy.

Clandestine production. Finally, there is the challenge of detecting clandestine reprocessing or enrichment activities. This is a challenge that is already faced in non-weapon states that are parties to the NPT. Iraq mounted a clandestine enrichment program as did Libya and Iran. In all three cases, the programs were discovered before they went into operation. For Iraq, the discovery was as a result of that country having to accept intrusive inspections after its defeat in the 1991 Gulf War. This helped lay the basis for the Additional Protocol under which non-weapon states commit to declare to the IAEA all significant nuclear-related activities and allow the IAEA to check those declarations. Iran voluntarily complied with the Additional Protocol for two and a half years between 2003 and 2006. During that period, the IAEA was able to visit suspect sites and detected undeclared enrichment-related activities.

The Additional Protocol also creates the possibility that the IAEA, if authorized by the IAEA Board, could carry out wide-area environmental monitoring to detect evidence of clandestine reprocessing or enrichment. There is a long Cold War history of atmospheric measurements of the concentration of the 11-year half-life fission product krypton-85 to detect foreign reprocessing activities. Recently, published analyses have begun to appear. Figure 7.3 shows the detection of krypton-85 at a site 60 kilometers away from a Japanese pilot reprocessing plant. Krypton-85 releases were detected with at least 50-percent probability down to levels corresponding to the separation of about 2 kg of weapon-grade plutonium per week, i.e., about one bomb equivalent per month.

Figure 7.3

The gaseous releases from centrifuge enrichment plants are very small. The uranium hexafluoride (UF6) gas in the system is at less than atmospheric pressure with the result that leakage is generally of air into the system rather than UF6 outward except when natural-uranium feed and enriched-uranium product cylinders are detached from the system. Air filtration systems are also standard equipment. Still, the degradation products of UF6 in the environment, molecules containing both uranium and fluorine, do not occur naturally. It is therefore worthwhile to determine if extremely sensitive detection techniques could be developed for such molecules. Furthermore, if tight controls could be established on UF6 at declared production plants, then a clandestine enrichment plant would require a clandestine UF6 production plant. Such plants produce the UF6 at above atmospheric pressure and therefore leak more UF6 than centrifuge enrichment plants.

When there is an indication of possible clandestine reprocessing or enrichment activity, the IAEA has the right to request an inspection. In a non-weapon state—and presumably in a nuclear-weapon-free world—inspectors would be free to take and analyze swipes. During the transition, at military nuclear sites in a weapon state, however, swipes could reveal information that a state considers sensitive: the isotopic makeup of or alloying material used in its weapon-grade plutonium, for example.

This is a familiar situation for the verification of the Chemical Weapons Convention (CWC) since chemical manufacturers wish to protect proprietary processes. Nevertheless, the Organization for the Prohibition of Chemical Weapons (OPCW), which is responsible for the verification of the CWC, uses sensitive instruments, notably gas-chromatograph mass spectrometers (GCMS) that are capable of identifying millions of chemical species and could be used for industrial espionage. For purposes of verifying the CWC, however, the chemical manufacturers and the OPCW have devised a “managed access” approach under which the library of chemical signatures inside the GCMS memory is purged of all information other than that relating to chemical-weapon agents, their precursors and degradation products.

The IAEA could similarly use instruments that have been rendered incapable of detecting anything beyond information required by the inspectors. Figure 7.4 shows, for example, a technique that could be used as a substitute for swipe samples. Laser breakdown spectroscopy could be used to turn particles on a surface into ionized plasma that would emit light with wavelengths characteristic of the particles’ constituent atoms. If spectral lines characteristic of uranium and fluorine were found together, that would be an indicator of gas centrifuge enrichment. The lines of all other elements could be blocked.

Figure 7.4

Thus, under an FM(C)T, the safeguards obligations of the nuclear-weapon states and the non-weapon states would begin to converge, with the IAEA having the responsibility of verifying non-production of fissile materials for weapons at both declared and suspect nuclear sites in all states. The authority of the IAEA to check for undeclared nuclear activities has been strengthened and codified in the Additional Protocol. It will be critical to the verifiability of nuclear disarmament that both weapon and non-weapon states ratify this Protocol.

In a nuclear-weapon-free world, several of the verification problems that will have to be dealt with today under a fissile cutoff treaty would be considerably eased. For one, there would be no stocks of fissile material not under international safeguards. Secondly, all states, including the nuclear weapon states, would have to adhere to a strict and strengthened Additional Protocol. Finally, managed access procedures could be greatly simplified because the nuclear-weapon states would no longer need to protect nuclear weapon-design information.