Editor's Note: On Jan. 18, diplomats from 65 countries at the U.N. Conference on Disarmament in Geneva will try to agree to begin talks on a treaty that would stop the production of plutonium and highly enriched uranium for nuclear weapons. Cutting off production of fissile materials for weapons could provide the basis for irreversible reductions of all nuclear weapons.

In the article below, Alexander Glaser, Zia Mian and Frank von Hippel, all at Princeton University and affiliated with the International Panel on Fissile Materials, propose the basic pillars of a fissile material cutoff treaty, how compliance with such a treaty could be verified and who should pay the tab. We'd like your views, which could help inform ongoing discussions of what should be in this treaty and how it might work, as well as help us further develop the article; please use the Comment feature at the end of this page.

Here are some questions to get you started:

Which arguments do you agree or disagree with? Why? Have the authors overlooked factors that could make a treaty more likely, less likely or impossible?

Should a treaty apply just to future production of fissile materials for weapons or should it try to capture the large stocks of unsafeguarded fissile materials that already exist but are not assigned to weapons?

What should be done with fissile materials in nuclear weapons that are no longer needed?

Should controls and inspections of fissile materials apply equally to all nations, or should nuclear weapon states be treated differently, even for their civilian nuclear materials?

Who should pay for the inspections to verify treaty compliance?

Treaties are already in place to cap the number of nuclear warheads that the U.S. and Russia can have ready to launch on missiles and bombers. Negotiations begun in 2009 could further lower each nation's ceiling, to less than 1,675 deployed weapons apiece. Separately, the U.K. and France have each reduced their nuclear arsenals to about200 and 300 warheads, respectively. China has a similar number, and Israel has about 100. But no agreements exist today that would effectively stop a country producing more nuclear material for weapons. Indeed, India, Pakistan and North Korea are increasing their weapon stocks. The world needs a verifiable treaty that would help cap the buildup of nuclear weapons and make reductions irreversible everywhere.

In May 2009 the United Nations' Conference on Disarmament agreed to hold talks on an international treaty to ban the production of fissile materials for nuclear weapons, generally referred to as a fissile material cutoff treaty. Fissile materials—in practice, plutonium and highly enriched uranium—are the key ingredients in sustaining the fission chain-reaction in nuclear explosives.

The talks, which participants hope will begin this spring, are long overdue. An end to the production of fissile materials for weapons purposes was first proposed in the early 1950s, but progress was stymied for decades by the Cold War. In 1993 the U.N. General Assembly agreed without dissent that there should be negotiations on such a treaty, but delays ensued because of drawn-out disputes over what such a treaty should cover, differences over priorities for the U.N. Conference on Disarmament and the George W. Bush Administration's skepticism toward arms control. President Barack Obama's April 2009 speech in Prague helped reopen the door to progress. He declared that, as one of the "concrete steps towards a world without nuclear weapons," his Administration "will seek a new treaty that verifiably ends the production of fissile materials intended for use in state nuclear weapons."

A cutoff treaty would make irreversible the moratoria in the first five nuclear-weapon states and end production in the others. It would strengthen resolve among non-weapon states for collective action against challenges such as those currently being mounted by North Korea and Iran. And it would give all nations greater grounds for insisting that the weapon states reduce their stocks of weapons-grade materials. At a minimum, those states could declare that existing stocks outside weapons will never be used for weapons. Pakistan has already made this call because India has a large stockpile of separated plutonium; although 90 percent of it is intended to be used as start-up fuel for its plutonium breeder reactors, all of it could be used for weapons.

To help inform the U.N. Conference on Disarmament negotiations, we propose here what a Fissile Material Cutoff Treaty (FMCT) should require and how compliance could be verified, and consider whether monitoring would be affordable.

No More Bomb Material
The most common fissile materials used in nuclear weapons are plutonium and uranium that is highly enriched in the isotope uranium-235 (HEU). About one kilogram of material—the amount fissioned in both the Hiroshima and Nagasaki bombs—releases an energy equivalent to the explosion of about 18 thousand tons of chemical high explosives. The International Atomic Energy Agency (IAEA) defines "significant quantities" of fissile material to be the amounts required to make a first-generation implosion bomb of the Nagasaki-type, including production losses. The significant quantities are eight kilograms of plutonium and 25 kg of U-235 contained in HEU. The U.S. has declassified the fact that four kg of plutonium is sufficient to make a modern nuclear explosive device.

As far as we know, the first five nuclear weapon states—the U.S., Russia, U.K., France and China—all stopped producing fissile materials for weapons before the end of the Cold War or soon thereafter. But they have not formalized their respective moratoria and no on-site verification arrangements exist. India and Pakistan are the principal countries building up today. Pakistan is investing heavily in new production facilities to try to keep up in its nuclear arms race with India. Israel continues to operate its 50-year-old plutonium-production reactor at Dimona, most likely to replenish the tritium used to boost the yield of its weapons (tritium has a 12-year half-life). North Korea had suspended production but claims to have started again in 2009.

The Treaty's Key Points
The three of us writing this article are involved with the International Panel on Fissile Materials (IPFM), a multi-national working group established in 2006 that brings together independent analysts and experts from numerous states. In November the panel published the Global Fissile Material Report 2009: The Path to Nuclear Disarmament, which proposes an approach to a FMCT and its verification that the U.N. Conference on Disarmament is considering in its January negotiations. Included are the basic tent poles of a treaty, which have been presented to diplomats and non-governmental experts and are being studied by various governments. The basic provisions are:

1. Each State party undertakes not to produce, acquire or transfer fissile material for nuclear weapons or other nuclear explosive devices.

2. Each State party undertakes either to promptly disable and decommission and, when feasible, dismantle its fissile-material production facilities, or to reconfigure and use these facilities only for peaceful or military non-explosive purposes.

3. Each State party undertakes not to use for nuclear weapons or other nuclear-explosive devices: fissile materials in its civilian nuclear sector, or declared as excess for all military purposes, or declared for use in military reactors.

4. Each State party undertakes that any reduction in its stockpile of nuclear weapons will result in a declaration of the fissile material recovered from those weapons as excess for weapon purposes.

5. Each State party undertakes to accept IAEA safeguards to verify these obligations.

Stockpiles Cut Significantly
These provisions, if passed, would have far-reaching effects. Fissile material stockpiles are huge. For example, about half of the separated plutonium in the weapon states was not produced for military purposes but is weapon-usable. Declaring it civilian and submitting it to IAEA safeguards would significantly reduce concerns that it might be used for weapons. Indeed, we have been told authoritatively that Pakistan's concerns about India's large stockpile of power-reactor plutonium—enough for about 1,000 warheads—would be much reduced if the material was declared civilian and placed under IAEA safeguards.

Russia and the U.S. have each declared that about 40 percent of their Cold War stockpiles of HEU, and roughly 25 and 50 percent, respectively, of their weapon-grade plutonium is unneeded for military purposes. Much of the excess HEU has already been eliminated but more remains; the proposed treaty would make its non-weapon status irreversible. After Russia and the U.S. reduce their stocks to sizes more comparable to those of other nations, those states would be expected to start placing their excess stocks under IAEA safeguards, as well. France and the U.K. could today declare as excess the fissile material in weapons they have removed from their arsenals since the Cold War.

The U.S. also has set aside 128 tons of its excess weapon-grade uranium—enough to make more than 5,000 warheads—as a reserve for future use in nuclear-power reactors for Navy submarines and ships. At current rates of use, this stockpile would sustain naval reactors for about 60 years. A few other weapon states have similar reserves, although they have not yet declared them publicly. Our proposed treaty would place this material under IAEA monitoring, too.

Verifying Compliance
Whenever arms-control agreements are discussed, much of the debate quickly focuses on whether parties to a treaty can make sure that other parties are not cheating. The Bush Administration was skeptical about the feasibility of detecting cheating and also was not eager to have IAEA inspectors knocking on the doors of sensitive U.S. facilities. It therefore argued for a treaty without inspections, except those that a country could mount unilaterally using imaging satellites and other "national technical means."

This position was not acceptable to the non-nuclear-weapon states that, under the existing Nonproliferation Treaty (NPT), are subject to IAEA inspections to verify that they are neither producing fissile materials for nuclear weapons nor diverting material from existing nuclear-fuel stocks for that purpose. IAEA verification of the NPT is far from perfect but the U.S. and other countries have been calling for strengthening it, not scrapping it.

The key challenges of verifying a cutoff treaty are the same as those for the NPT: verification that HEU produced in declared enrichment facilities and plutonium separated in declared reprocessing facilities is placed under IAEA safeguards; detection of any undeclared production activities; and verification that fissile material—whether newly produced or pre-existing—is not removed from international safeguards once it has been submitted to IAEA monitoring.

Verification of the FMCT in the civilian sectors of the weapon states could be based mostly on the same procedures developed by the IAEA for verifying the NPT in the non-weapon states. A few additional challenges would arise, however, in weapon states where plutonium is separated from spent power-reactor fuel at reprocessing plants and uranium fuel is enriched in gas centrifuge plants that previously were used for producing weapons materials.

One problem is that the IAEA would not be able to fully check the designs of pre-existing facilities in the weapon states. This would be especially problematic for reprocessing plants, where many pipes are buried in concrete and some areas are inaccessible because of high radiation levels. But the IAEA could focus on checking that a plant functions as declared. After allowing for expected losses, the amount of separated plutonium measured during annual plant cleanouts should match the amount measured when the spent fuel was dissolved. Unlike an accountant's check of the flow of money through a bank, however, there would be measurement errors. In a large reprocessing plant such as that at La Hague in France, which can separate more than 10 tons of plutonium a year, these errors could conceal the diversion of enough material to make on the order of 10 nuclear weapons a year. Fortunately, reprocessing is both unnecessary and uneconomic and persists in only a few countries for bureaucratic and political reasons. Among the non-weapon states, only Japan reprocesses today. Nuclear-energy use could continue, even on a much larger scale, for many decades without reprocessing and separation of plutonium from spent fuel.

A Switch to Low-Enriched Uranium
Countries also have little reason to produce HEU. Power reactors are fueled with low-enriched or natural uranium and HEU-fueled research reactors are being phased out because of concerns about theft of the uranium. Increasingly, only naval propulsion reactors will be using HEU for fuel. Yet, for economic reasons, France's new generation of submarine and aircraft carrier reactors are fueled by low-enriched uranium (LEU). It is not known whether China uses HEU or LEU. The U.S., Russia and the U.K. use HEU but have large reserves of excess weapon HEU that could sustain their naval reactors for many decades. Only India is known to be producing HEU for its naval-reactor program. Given concerns that large stocks of HEU would raise in a world moving toward much smaller weapon stockpiles, all countries with nuclear navies should move to LEU fuel.

In the huge enrichment plants that produce most LEU for power-reactor fuel, the IAEA has a powerful technique to detect clandestine production of even small quantities of HEU. It takes "swipes" with gauze pads of the walls, machines and piping in a facility and then scans the pads back at its laboratories for microscopic particles of any HEU that condensed from leaks of uranium-hexafluoride gas, used during the enrichment process. Such swipes revealed undeclared enrichment activities in Iran.

One complication with swipes is that some of Russia's huge centrifuge enrichment plants that today make LEU had once produced HEU. But that was more than 20 years ago and it appears possible to distinguish large particles of old HEU from any new particles that might be illegally created after a cutoff treaty comes into force, for example by measuring the fraction of U-234 that has decayed into thorium-230 and by other features of its minor-uranium-isotope "fingerprint."

Detecting Clandestine Activities
Under the FMCT, the problem of detecting clandestine enrichment or reprocessing would be basically the same as the current challenge the IAEA faces in detecting clandestine production in non-weapon states that have signed the NPT.

After Iraq's secret nuclear-weapons program was discovered in 1991, the IAEA, with help from member countries, began to work on improved methods to detect such activities. One result was the Additional Protocol, under which countries agree to give the IAEA much more information about and access to their nuclear-related sites—including the ability to monitor the air, soil, water, plant matter and surfaces outside of facilities to detect the release of telltale radioactive gases or other effluents.

If sample analysis offers a solid basis for suspecting clandestine enrichment or reprocessing activities, the IAEA can request an on-site inspection. If unrelated sensitive information exists at a site, the host country has the right to conceal that information as long as its concealment measures do not interfere with the IAEA's ability to detect production activities.

If the IAEA obtained credible evidence that clandestine production might be taking place at a facility associated with nuclear weapons or naval propulsion reactors, "managed access" arrangements with the site operators would be required to assure that the inspectors could detect key indicators of enrichment or reprocessing without collecting data that would reveal sensitive information about nuclear-weapon or naval-fuel design. Enrichment and reprocessing plants would not be easy to conceal from inspectors with access to the site. They have characteristic physical features and unique radiation and chemical indicators.

The Neutron-Beam Test
As Russia and the U.S. have reduced their arsenals, they have declared as excess some of the HEU and plutonium that was in these weapons. For a time, both countries were interested in demonstrating the irreversibility of their reductions by offering to have the IAEA monitor this material even while it was still in weapon components. Under the FMCT, this process could be formalized to cover the material in all weapons that are cut from arsenals. This would ensure the material is not covertly returned to weapons use.

In the 1990s Russia's Ministry of Atomic Energy and the U.S. Department of Energy joined with the IAEA in a "Trilateral Initiative" to show that it is possible for the IAEA to verify that a closed container holds weapon-grade plutonium, while keeping design details of the weapon component secret. Radiation detectors would measure the gamma and neutron radiation coming from the container and a computer would analyze the measurements. Classified information then would be filtered out by a simple computerized "information barrier" and the inspectors would only see the results of the analysis through three pairs of red and green lights. The green or red light in each pair would illuminate depending upon whether or not the container held plutonium, the plutonium was weapon-grade and more than a specified amount of the plutonium existed inside.

The IPFM has been investigating a similar approach for weapon components or naval-reactor fuel containing HEU. Since HEU emits gamma rays and neutrons at a much lower rate than plutonium, external neutron beams would be sent into the container to induce fission reactions, which would create signature neutron and gamma emissions.

Perhaps the most challenging verification task for our proposed fissile material cutoff treaty—politically, at least—would be to determine that no HEU was being diverted from the fuel cycles of naval-propulsion reactors. At least some countries, including the U.S., consider fuel and reactor-design information to be sensitive.

As already noted, the easiest solution to this problem would be for all nuclear navies to design their future reactors to use low-enriched uranium—as France already does. But HEU-fueled reactors will be with us for some time. We believe that a combination of the Trilateral Initiative approach and some limited facility-perimeter control to assure that HEU is not removed at some stages in the core fabrication and installation processes could provide a solution. Oak Ridge National Laboratory is developing a technique using neutrons to interrogate HEU-containing objects to detect diversion of HEU from small reactor cores.

Cost, and Who Pays
The IAEA would be a natural choice to verify the FMCT. It has extensive experience in inspecting nuclear facilities in the non-weapon states, and some experience in the nuclear weapon states. To undertake the additional responsibilities required under an FMCT, we estimate that the IAEA might have to double its inspection budget, which is currently about $100 million a year.

The typical cost of safeguarding a nuclear facility is expressed in annual person-days of inspection (PDI). The IAEA conducts about 10,000 PDI a year in the non-weapon states, yielding a total cost—including administrative overhead, travel and special laboratories to analyze samples—of about $10,000 per PDI. A typical light-water power reactor requires about 10 PDI per year. A large commercial centrifuge facility requires about 150 PDI a year.

Spent power-reactor fuel reprocessing plants are far more challenging and costly to safeguard. Japan's Rokkasho reprocessing plant, the only one in a non-weapon state, and custom-built to be safeguard-friendly, requires about 1,000 PDI per year to monitor. Newer verification and monitoring tools and methods could bring the inspection effort down to about 200 PDI per year. Verifying an FMCT will become easier, less expensive and more effective as the remaining older reprocessing facilities are shut down, dismantled and (hopefully) not replaced.

We offer three options in our draft treaty for who might pay the additional costs for FMCT verification. Of course, the diplomats negotiating the treaty would have to decide which option they think is best:

All IAEA member states would increase their contributions to the IAEA. This is how the NPT is funded today.

All countries that are party to the FMCT would pay the additional costs according to a formula that reflects the relative sizes of their gross national products. This is how contributions to the UN are set and is typically how international treaties are funded.

Only nuclear-weapon states would pay, because this is where verifications would have to occur. One formula might take into account a combination of the UN formula and the capacities of each country's enrichment and reprocessing plants.

As we have crafted the main points of a proposed treaty, we and other people in the IPFM have concluded that the technical challenges to achieving a verified FMCT are manageable. But political challenges would have to be overcome, too. Countries that are still building up their nuclear-weapon stocks—especially India and Pakistan—would have to decide that they have enough fissile material. And countries and nuclear establishments within them that are reluctant to accept international inspections, including China, Russia and the U.S. Navy, would have to be willing to cooperate. A verified FMCT is an essential first step toward creating a system of international controls on nuclear materials that will allow nations to cut their weapon stockpiles deeply and move toward a world without nuclear weapons.