Plutonium End Game

Press Release

Table of Contents

Preface

Summary and Recommendations

Chapter One: Nature of the problem of commercial plutonium

Chapter Two: A Brief History of Commercial Plutonium

Chapter Three: Assessment of the current situation

Chapter Four: Disposition of US-Russian Surplus Military Plutonium

Chapter Five: Alternative Disposition Options

References

Having failed in the commercial sector, the plutonium fuel industry may have found a crutch in the military sector. The end of the Cold War raised the issue of surplus plutonium from dismantled nuclear warheads that were no longer required for military purposes by the United States and the Soviet Union/Russia. In turn, the issue of the disposition of surplus plutonium is part of the larger issue of the security of nuclear weapons and weapons-usable materials that achieved some urgency because of economic distress in Russia in the 1990s and the frequent political changes there.

The attempted coup in August 1991 in the Soviet Union dramatized the problem. For several days, no one knew who, if anyone, had operational control of the Soviet nuclear arsenal. Within weeks, President Bush unilaterally withdrew most US tactical nuclear weapons from deployment and from the US arsenal altogether, hoping that President Gorbachev would reciprocate. He did, in about a week. In the same period, the United States and the Soviet Union arrived at the START I agreement to reduce their strategic nuclear arsenals. The dismantlement in both Russia and the United States of thousands of nuclear weapons has occurred simultaneously with the decline in the availability of financial resources to Russian nuclear weapons production and design centers. While the problem of diversion of nuclear weapons materials into black markets has always been a serious security concern, the issue took on greater importance and urgency with the disintegration of the Soviet Union and the decline of its economy.

Between them, the United States and Russia have declared about 100 metric tons of plutonium surplus to their military requirements, of a total stock (in various forms) of about 220 metric tons. Further, the Soviet Union also operates a commercial reprocessing plant, RT-1 within its Mayak nuclear weapons facility near the city of Chelyabinsk in the southern Urals. Commercial spent fuel reprocessing over the last two-and-half decades has created a stock of about 30 metric tons of plutonium that is stored on the Mayak site. Soviet plans, like those in France and Japan, and earlier hopes in the United States, called for an electricity sector based on breeder reactors. But, the two large-scale breeder reactors in the Soviet Union (one is in Russia, the other in Kazakhstan) operate on medium enriched uranium fuel.⁵⁰ By the end of the 1980s, Soviet plans for building additional breeder reactors stalled for lack of money and due to increasing public opposition.

In contrast to several western countries and Japan, Russia's nuclear energy ministry Minatom, and its Soviet predecessor did not make plans to use plutonium in the form of MOX fuel in light water reactors during the 1980s. This is because such use degrades the isotopic composition of plutonium, creating increasing amounts of the higher isotopes, plutonium-240, -241, and -242, with each pass through a light water reactor. The presence of increasing amounts of these higher plutonium isotopes makes the plutonium unfit for repeated use as fuel. Moreover, it is uneconomical to reprocess such MOX spent fuel just to re-extract the uranium, since it has little value. Indeed, the uranium recovered from reprocessing plants is increasingly coming to be viewed as an economic liability since it is contaminated with traces of fission products, plutonium, and neptunium as well as undesirable isotopes of uranium that are created in the reactor, uranium-232 and uranium-236.

These technical factors mean that MOX use in light water reactors cannot convert, even in theory, more than a few percent of the uranium-238 resource base into plutonium. By contrast, the use of MOX in fast breeder reactors does not degrade the isotopic composition of the plutonium, since, unlike the thermal neutrons in a light water reactor, the fast neutrons in a breeder reactor also fission the non-fissile isotopes of plutonium, namely plutonium-240 and -242.⁵¹ Moreover, since plutonium-239 of relatively high purity can be made in the breeder reactor "blanket" by putting uranium-238 there, breeder reactors can actually be used to improve the isotopic composition of plutonium (in the sense of increasing the percentage of plutonium-239).

These technical factors underlie Minatom's goal of developing an energy economy in which plutonium-fueled breeder reactors would play a major role. Hence, immobilizing plutonium and treating it as a waste has not been well-regarded by Minatom in its plutonium disposition negotiations with the United States.

From a purely technical point of view, an electricity system based on nuclear reactors that aims to use the uranium resource to the fullest must be based on breeder reactors, and not on the use of MOX in light water reactors. However, the use of most of the uranium-238 resource base, while it may be technically appealing to some, is not necessarily a rational economic goal. The merits of a narrow physical resource-related goal must be compared to other available options in terms of resource use, cost, performance, and environmental criteria, even aside from the safety or non-proliferation related considerations that are unique to plutonium fuel.

The lack of a MOX fuel fabrication plant in Russia and the stalling of its breeder reactor program means that essentially all its commercial separated plutonium is stored, unutilized (a situation similar to that in Britain). Further, while Russia has some experience at the pilot plant level in MOX fuel fabrication, there is no commercial plant for making MOX fuel in Russia. In the absence of a significant number of breeder reactors, the Soviet Union did not build a commercial-scale MOX fuel fabrication facility. Russia is producing small amounts of MOX fuel at a pilot plant in Dmitrovgrad for a small breeder reactor, BOR-60, located there.

The disposition of surplus military plutonium in Russia therefore comes, for Minatom, within the larger context of its own plans and ambitions for a plutonium economy based on breeder reactors. The US disposition approach is more complex. Its fundamental aim has been to create a program that, in the shortest feasible time, would permanently put as large a portion of Russian military plutonium as possible into non-weapons usable forms that cannot be diverted easily. In order to do that, the United States realized that it must propose a program that would also put its own surplus plutonium into non-weapons usable form(s).

The evolution of US-Russian proposals to deal with surplus military plutonium must be seen in light of these conflicting objectives. If Russia were to attain its goal of a plutonium economy, then construction of a MOX plant suitable for breeder reactors would be a principal objective, along with the reprocessing of spent MOX fuel. The US goal meant that spent MOX fuel should not be reprocessed, for to do so would result in the re-separation of weapons usable material. Further, since separated commercial plutonium is also weapons-usable, the same underlying tension applies to any disposition strategy for Russian separated plutonium.

In 1994, the US National Academy of Sciences proposed a two-track approach to plutonium disposition that sought to strike a balance between the US government and Minatom goals. A part of the surplus military plutonium would be fabricated into MOX fuel for light water reactors and used in a once-through mode as a fuel. This would extract some of the fuel value from the plutonium, though at a net cost, since plutonium is far more expensive to fabricate into a fuel that uranium. And a once-through mode of use, that is, without reprocessing, would mean that the plutonium fabricated into MOX would be permanently put into a non-weapons usable form.

This was so far from Minatom's objectives that subsequent negotiations have altered the plan in fundamental ways towards Minatom's goals. The June 4, 2000 US-Russian agreement, which was signed by Vice-President Gore on September 1, 2000,⁵² provides for the disposition of 34 metric tons of surplus weapons grade plutonium for each country in the following way:⁵³

• MOX fuel plants would be built in Russia and the United States.
• Russia would convert all of its surplus weapons plutonium into MOX fuel, while the United States would use 25.5 metric tons as MOX and immobilize 8.5 metric tons in a ceramic matrix that would then be put into highly radioactive glass logs to provide proliferation resistance.
• At least 2 metric tons per year would be dispositioned, with provision to increase this by another 2 metric tons per year. Hence, once the plants are built (the target date for which is currently 2007), it would take between 8.5 and 17 years for each side to put its 34 metric tons into non-weapons usable form.
• Russia would be allowed to reprocess MOX spent fuel after the 34 metric tons has been used in reactors under international safeguards.
• The West would pay for the costs of the plutonium disposition program in Russia, though not all costs would be covered (see below)
• Future surplus plutonium would be dealt with along the same lines.

However, Russia and the United States have not been able to arrive an agreement on who would bear the liabilities arising from the program. A central question in this regard is the question of who would pay the damages to third parties, such as European countries, in case a severe reactor accident on the scale of Chernobyl, were to spew radioactivity over Europe.

The United States and Russia lack only a MOX fuel fabrication plant and some related processing facilities to complete the infrastructure for MOX fuel production. Once that is done, there will be an incentive to continue plutonium separation, which would entrench plutonium use. And, as noted, the US-Russian agreement allows Russia to reprocess MOX spent fuel, after a delay, which may be as short as ten to fifteen years. The US-Russian MOX program will increase proliferation and safety concerns, while also increasing the cost for disposition.

Russia clearly sees the weapons disposition program as a way to establish a plutonium fuel economy. According to Minatom:⁵⁴

"Disposition of weapons plutonium must be seen as the first step in developing a technology for a future closed nuclear fuel cycle. The basic direction in the disposition of excess weapons plutonium, as with plutonium from spent nuclear fuel, is the use of mixed uranium-plutonium fuel of fast reactors, which forms the basis for future large-scale nuclear power engineering. The disposition of a limited amount of weapons plutonium in thermal reactors, if this requires political approval, can be carried out under the financial and technological cooperation of the world community."

In other words, Minatom does not appear to care whether MOX fuel use in thermal reactors (including light water reactors) is carried out abroad or in Russia, so long as the West is the responsible party for the program. Recently, Atomic Energy Minister Yevgeny Adamov stated that the cost of modifying Russian light water reactors to use MOX would be high because Russia does not use MOX in these reactors. He suggested that Russia would consider selling MOX fabricated in a Russian plant to other countries for irradiation there.⁵⁵

In sum, Minatom's plans make it abundantly clear that its core program is to build, at Western expense, the MOX fuel fabrication infrastructure for breeder reactors. Minatom also hopes to build new breeder reactors at its own expense. Minatom may also want to make some money by selling MOX made in Russia to third countries for irradiation in thermal reactors. However, given the immense and growing surplus of commercial plutonium, it may be difficult to find customers for Russian MOX fuel.

A MOX fuel program would create the infrastructure of facilities and financial interests for continuing "commercial" plutonium use. Once begun for weapons plutonium, the MOX plant is likely to be used for plutonium separated from commercial spent fuel. This intent is explicit in Russia, and currently denied in the United States. But once the plant is there, it will be politically difficult to shut it. As evidence, we point to the fact that two military reprocessing plants are still operating at Savannah River Site in South Carolina, supposedly for reasons of "environmental management." In the same way, Russia is also operating two military reprocessing plants, one at Tomsk-7 and the other at Krasnoyarsk-26, as a way of managing its spent fuel. In the process, Minatom continues to inject highly radioactive liquid wastes deep underground. Low-level liquid radioactive wastes are discharged into aquatic environment.

A MOX program would put plutonium on the roads and in commercial nuclear power plants. At every step of the MOX process until it is loaded into a reactor and its irradiation has begun, the plutonium would need to be safeguarded with a military level of security. This is because plutonium can be separated from the uranium in MOX fuel by relatively simple steps and facilities compared to reprocessing spent fuel. The MOX production facilities, transport to power plants, and its storage at power plants would all be points of potential theft. Unless steps are taken to militarize security at these facilities, there will be a risk of diversion of weapons-usable, including weapon-grade, plutonium.

Proponents of the current US-Russian MOX plans argue that a moratorium on reprocessing of MOX spent fuel during the disposition program would be a sufficient safeguard against proliferation. But it is irrelevant whether MOX spent fuel is reprocessed now or in one or two decades. There is plenty of other spent fuel to reprocess now. This can keep reprocessing plants going until the moratorium on MOX spent fuel reprocessing expires. MOX spent fuel can conveniently be put at the end of the line. Moreover, the US-Russian plan does not address the 30 metric tons of separated commercial plutonium stored at Mayak in the southern Urals. So the net result will be that first military plutonium will be used in the MOX plant, decreasing the military plutonium stock, while commercial reprocessing increases the commercial plutonium stock. Then the military-origin MOX spent fuel can be reprocessed, while the already separated commercial plutonium is fabricated into MOX fuel. In the meantime, new breeder reactors would be built. All but the last element of the plan would be financed with western money. This seems to be the plan the Minatom is banking on.

The vast majority of light water reactors were not designed to use plutonium as a fuel. Several specific differences between MOX fuel and low-enriched uranium fuel affect safety:

• The rate of fission of plutonium tends to increase with temperature. This is called a positive temperature coefficient of reactivity and can adversely affect reactor control.
• Reactor control depends on the small fraction of neutrons emitted in the seconds to minutes following fission of uranium or plutonium. These neutrons are called delayed neutrons. On average, plutonium yields about three times fewer delayed neutrons per fission than uranium. This means that provision must be made for increased reactor control if MOX is used. Otherwise, the reactor would become less safe. How many more control elements are needed, where they must be placed and whether any particular reactor can be modified to accommodate MOX fuel, are questions that must be examined with every specific light water reactor design. For instance, only one of the three designs of pressurized water reactors that are being used in France for power production can be modified to accept MOX fuel.

• MOX fuel also causes more irradiation damage to the reactors than traditional fuel, because the neutrons resulting from MOX fuel use in light water reactors have a greater average energy than those resulting from uranium fuel. This is due to differences in the fission cross sections between plutonium and uranium in various parts of the thermal neutron spectrum.
• Irradiated MOX fuel is thermally hotter than uranium fuel. Further the larger quantities of transuranic elements produced during reactor operation with MOX fuel, complicate repository disposal of MOX spent fuel.

The larger quantity of transuranic elements, like americium, present in irradiated MOX fuel means that the consequences of a serious accident would be even more severe than for a reactor fueled with low-enriched uranium. According to a report by the Nuclear Control Institute in Washington, DC, the public health consequences of severe accidents involving light water reactors with MOX cores are likely to be considerably greater than those involving low-enriched uranium (LEU) cores.⁵⁶ This is mainly because the higher proportion of plutonium in the fuel would increase the release of plutonium and other transuranic elements to the environment in case of a severe accident.

Historically, VVER-1000 reactors have experienced significant problems in the past, just using the uranium fuel for which they were designed. Vladimir Kuznetsov, a reactor expert, who worked for Gosatomnadzor for many years, Russia's nuclear regulatory agency, has expressed serious concerns about VVER-1000 reactor safety and published a book on the subject.⁵⁷ Use of MOX fuel in VVER-1000 reactors would complicate the safety issues associated with the operation of these reactors. It would become more complex to assess the conditions of their safe operation, for example.

The fact that Minatom is not particularly interested in the use of MOX fuel in light water reactors is a further problem. As discussed above, Minatom's clear and publicly stated goal is to use the weapons plutonium disposition program to further its goal of creating a long-term program for the use of MOX fuel in breeder reactors. This has become an important part of its goal of creating a plutonium-based electricity sector.

Another safety issue specific to weapons grade plutonium is that it has never been used as a commercial fuel. Reactor-grade plutonium has different isotopic characteristics than weapon-grade plutonium. Reactor grade plutonium from light water reactors has roughly 60 percent plutonium-239. Weapon-grade plutonium has about 94 percent plutonium-239. The US Department of Energy is proposing 40 percent MOX fuel in the reactor core,⁵⁸ compared to the 30 percent that is normal in the reactors which use MOX fuel in France. This means that in US reactors fueled with MOX, there will be about 38 percent plutonium-239 as fuel, compared to about 20 percent for commercial reactors, which represents almost a doubling of the amount of plutonium-239 in the reactor core compared to French experience. Despite this crucial difference, the US Department of Energy intends to use the computer codes from French reactors using commercial MOX fuel to analyze weapons plutonium MOX performance in reactors. Such large extrapolations from commercial experience will create significant uncertainties about safety.

In Russia, the pressure to increase the MOX fuel loading of the core will be even greater. Achieving the 2 metric tons per year rate disposition rate in Russia has been a concern, because the number of Russian light water reactors is limited. Moreover, all of the existing reactors that are under consideration for MOX will reach the end of their 30-year lifetimes within the next 25 years and many of them much earlier. The pressure to extend the lifetimes of the reactors will be very great. Other possible measures include using a large percentage of MOX in the BN-600 reactor core, using reactors in Ukraine, or using CANDU reactors in Canada. All of these approaches pose additional safety risks that have not been adequately addressed.

In the United States, the DOE may ask for a loading of MOX fuel of as much as 40 percent of the reactor core, even though the experience with MOX in France, whose computer codes would be used to evaluate US reactor operation, is with a 30 percent core. Further, the neutron economy of this 40 percent of the core and hence of the entire reactor will be different from that prevailing in French reactors using MOX. This is because commercial MOX contains reactor-grade plutonium, which has about 60 percent plutonium-239, whereas weapon-grade plutonium contains about 93 percent plutonium-239. The use of French computer codes for evaluating weapon-grade MOX fuel use therefore raises some procedural, technical and safety concerns.

In order to satisfy US demands, Minatom has agreed to use MOX in light water reactors, rather than breeder reactors as they would prefer. So far as we understand, Minatom had never seriously considered the use of MOX fuel in its light water reactors, until the United States brought the idea to the table as part of the military plutonium disposition plan. This raises some liability questions, such as who will pay for the costs, if an accident should occur in a light water reactor using MOX fuel.

The liability question has not been resolved in the September 2000 US-Russian agreement. This is a principal problem that could have serious implications. Russia does not possess the financial resources to insure the MOX fuel program. Were there to be an accident in a light water reactor using MOX fuel, Russia could, with reason, claim that this was a western idea implemented for western purposes with western money. Even the money for funding the Russian nuclear regulatory agency, Gosatomnadzor, is to come from the West. Hence, whatever agreement is reached about liability, the reality of the situation is that it will be borne by the people of the west and of all those areas in Russia and elsewhere that are at risk of contamination in case of an accident.

Given that the dangers of using MOX fuel in Russian light water reactors are being publicly discussed, parties injured as a result of accidents could reasonably claim that companies such as Cogéma, Siemens, and BNFL, which are all likely to be involved in the implementation of the agreement, knew of the risks but went ahead with the program anyway. Hence, the liability issues associated with the MOX program not only have significant financial implications for the US and Russian governments, but also for the people, governments and corporations of France, Britain, and possibly Belgium. Yet despite the grave safety and economic issues involved, there has been little public discussion of liability issues arising from the US-Russian plutonium disposition agreement.

An independent evaluation of whether the safety standards involving Russian MOX use would be comparable to those in the United States is needed. Even in the United States the regulatory situation in regard to military plutonium disposition is far from satisfactory. The US Nuclear Regulatory Commission has not exercised adequate oversight in the process of selection of which reactors might be suitable or not for MOX fuel use. In fact, it has allowed commercial interests to dominate the process by allowing any reactor owners to bid for MOX fuel contracts. A prudent process would have ruled out some reactors on the basis of design or other safety considerations. It would have allowed only those owners of nuclear power plants to express interest that passed a set of stringent safety and design criteria to express interest in the MOX program.

Further, the Russian regulatory agency Gosatomnadzor appears to have far less authority over actual decision-making in regard to licensing and operation of facilities than the Nuclear Regulatory Commission in the United States. Corresponding to its low political clout, Gosatomnadzor has little money and will likely depend on US money for regulating MOX fuel use. This is a serious vulnerability and shows the relatively weak position of this agency in the Russian nuclear establishment.

The estimated budget for the disposition of Russian military plutonium makes no provision for liability insurance. This implies that Russia will bear the costs, if an accident should occur in a light water reactor using MOX fuel. The state of the Russian economy is such that Russia would not be able to cover such costs. Even the US the provision for liability for its own reactors, $10 billion, is grossly inadequate. Even in better economic times, the resources of the Soviet Union were not adequate to the task of paying for the costs of the Chernobyl accident. Remembering the grave damage that many countries besides Ukraine, Belarus, and Russia have suffered as a result of the Chernobyl accident, the European people might reasonably ask whether how they would be compensated in case of an accident.

Even if the United States agrees to assume a portion of the liability, it is far from certain that the US Congress would actually provide the money, should it be needed. This could create its own set of political difficulties. In sum, the liability for use of MOX fuel in Russian light water reactors presents a very difficult problem. That would continue to be the case even the United States and Russia arrive at a formal agreement on the liability question, since it is highly unlikely that such an agreement could designate resources to compensate victims in case of a severe accident.

The current estimate is that the Russian MOX program will cost $1.7 billion⁵⁹ and the US program will cost about $4 billion.⁶⁰ These cost estimates do not include many items such as the cost of insurance or the costs of electricity while the reactors are shut down for modifications. These costs could be significant. Who will pay for them is still an unresolved issue. Even though plutonium will be used to generate electricity, the use of MOX fuel will involve net costs. This because it is more expensive to fabricate MOX fuel even when the military plutonium is free than it is to purchase low-enriched uranium fuel. In addition, the extra security, which will add to the cost of the whole process, has not been adequately considered in the evaluation of the MOX program.

The total amount of the net cost of the MOX plan is a matter of some debate. It will depend on the actual costs of processing weapons grade plutonium into MOX fuel, the costs of reactor modifications, the time it will take to make reactor modifications, the cost of the replacement electricity, and the cost of ensuring and certifying the safety of reactors for MOX fuel use. Any international transport of plutonium, for instance to Canada, would add to the costs and risks.

A considerable amount of discussion has revolved around the financing of the Russian portion of the plutonium disposition plan. In order for the MOX plan to be implemented, the wealthy governments of the West and Japan, called the Group of 7 (or G-7) must agree to finance it. While Russia and the United States are pressing Japan and the European member of the G-7 to help finance the plan, no money has as yet been commitment beyond US funding for the design of the Russian MOX plant.

It is also possible that private financing might be developed. One financing plan has been put forward by a US corporation known as the Non-Proliferation Trust, Inc. This would involve importation of up to 10,000 metric tons of foreign spent nuclear power reactor fuel for storage in Russia, a complete halt to commercial reprocessing, and a payment to Russia for building storage facilities, a nuclear waste repository and other purposes. Such a plan is currently illegal under Russian law and hence attempts are being made to amend the law. However, many Russian, US and other non-governmental environmental organizations oppose this plan.

The commercial plutonium industry is currently uneconomical and could survive only with vast subsidies from taxpayers and electricity ratepayers. The US-Russian disposition agreement would provide MOX fuel use with even more subsidies, which, in Russia's case, would be used to provide new life to its moribund breeder reactor program. This would renew hopes for a plutonium economy in the nuclear establishments not only in the United States and Russia but also in other countries, notably France, Germany, Japan, and possibly also Britain.

Next: Chapter Five: Alternative Disposition Options