Status of research and development after the IMF8 Workshop

Inert matrix fuel has the potential to produce electricity while burning up more plutonium

Claude A. Degueldre, a scientific project leader at the Paul Scherrer Institute (PSI) in Switzerland, briefed a Los Alamos National Laboratory audience March 20 on the status of research and development on inert matrix fuel (IMF)-an approach that has the potential to greatly reduce the amount of reactor-produced plutonium that should be placed in geological disposal.

Degueldre's host at the Seaborg Institute Seminar was David Clark, institute director, who noted that Degueldre, a Doctor és Sciences from the University of Liege, Belgium, and currently a professor at the University of Geneva, Switzerland, also gave a Seaborg Institute lecture two years ago.

Degueldre, who has expertise in the analytical chemistry of actinides as well as IMF, said that Switzerland has five light-water reactors. As a result, its stockpile of plutonium-a product of light-water reactors fueled by uranium-is increasing by approximately 750 kilograms per year. The Swiss population, he said, would like to get rid of this plutonium. "My work for the last 10 years was to find a scientific solution to that problem," he added.

Claude Degueldre

Countries including Japan and France are facing the same problem-one that is also important in the United States, a nation with much more "open space," where the current plan is to bury long-lived, high-level waste in Yucca Mountain near Las Vegas, Nev., and in similar future geologic depositories where it will have to remain for more than 10,000 years. Plutonium, a radiological hazard, has a half-life of 24,000 years, and the world's nuclear reactors are now producing about 70,000 kilograms of it per year.

Research indicates that IMF might provide some valuable answers to the problem, making it possible to produce electricity while burning up more plutonium and minor actinides in a way that is proliferation-resistant, economical, and ecological, while also being safe and sustainable.

Since 1995, Degueldre has been the primary organizer of a series of world conferences on IMF. The most recent, "IMF8," was held in Japan. Degueldre said in his Los Alamos presentation that the number of scientists working on IMF is growing. He said these scientists, from many different countries and various research fields, have become "a group of friends working together," evaluating a variety of strategies, exchanging samples and data, and comparing results at each annual meeting.

The process of producing plutonium isotopes in a nuclear reactor is shown here. Fertile ²³⁸U captures a neutron to produce ²³⁹U. The ²³⁹U is unstable with a half-life of 23.5 minutes and decays via loss of a beta particle to produce ²³⁹Np, which is also unstable toward beta decay, and generates ²³⁹Pu. ²³⁹Pu can either under fission to produce more energy, or it can capture neutrons to successively produce ²⁴⁰Pu, ²⁴¹Pu, etc.

In a recent paper-"Inert Matrix Fuel Strategies in the Nuclear Fuel Cycle: the Status of the Initiative Efforts at the 8th Inert Matrix Fuel Workshop"-Degueldre and a coauthor, T. Yamashita of the Japan Atomic Energy Research Institute (JAERI) in Tokai, Japan, explained the motivation for their studies and summarized the work in progress.

"The 'raison d'être' of the Initiative for Inert Matrix Fuel," the authors said, "is to contribute to research and development studies on inert matrix fuels that could be used to utilize, reduce, and dispose both weapon- and light-water-reactor-grade plutonium excesses."

They said IMF, a promising once-through strategy, could be used in the existing commercial light-water reactors in Europe, Japan, Russia, and the United States; in the Canadian pressurized heavy-water reactors; or in future transmutation units where actinides could be bombarded by neutrons and converted to different nuclides with shorter half lives.

"This option has the advantage of reducing the plutonium amounts and potentially minor actinide contents prior to geological disposal," Degueldre and Yamashita said in their paper.

On the other hand, IMF could be used in a "multirecycling strategy"-in other words, the actinides encapsulated in inert matrices could be run through a reactor or a transmutation device more than once; each time through, more would be consumed. After a last cycle, the final spent IMF would be disposed of in a geologic repository.

Plutonium production in UOX vs. plutonium consumption in IMF

In conventional nuclear reactor fuel, fissile ²³⁵U is enriched to approximately 3 percent in a matrix of 238U, usually in the form of a UO₂ (or UOX) fuel pellet.

While not directly fissionable, the ²³⁸U is fertile in the sense that it can capture a neutron to generate ²³⁹Pu, which is fissile, and the ²³⁹Pu can capture neutrons to generate ²⁴⁰Pu, ²⁴¹Pu, again fissile, etc. In this way, fertile ²³⁸U is transmuted into isotopes of plutonium (and other actinides) in the reactor neutron spectrum.

During the lifetime of the fuel in the reactor, some of this plutonium is also burned up in fission reactions, but a significant amount remains in the spent nuclear fuel, generating concern about the potential proliferation of plutonium. Spent fuel can be reprocessed to separate out the plutonium for burning in a reactor in the form of a mixed-oxide (MOX) fuel. While the process of burning separated plutonium in the form of MOX has a long history, the pace of plutonium removal is slow or unachievable because while MOX burns some plutonium, the neutron irradiation of the fertile ²³⁸UO₂ matrix continues to produce more plutonium.

Ideally, one would like to replace the fertile ²³⁸UO₂ matrix with a nonfertile, "inert matrix" to avoid the production of plutonium in nuclear reactors and to efficiently achieve its consumption. This is the concept of an inert matrix fuel.

IMF composition and loading

Degueldre described how the material selection for the inert matrix is guided by the neutronic properties of the elements and/or isotopes based on their transparency for neutrons in the reactorÑan essential requirement for the term "inert matrix."

The desired thermodynamic properties of the materials include high melting point (approximately 3,000 Kelvin), good thermal conductivity, chemical compatibility with cladding, low chemical solubility in hot water, and high density.

Detailed studies around the world have produced a number of candidate materials that include stabilized ceramics such as Ca_xZr_1-xO_2-x, Y_yZr_1-yO_2-y/2, or other ceramics such as ZrSiO₄, Y₃Al₅O₁₂ (yttrium aluminum garnet), MgO or MgAl₂O₄ (spinel), or even nitrides or carbides. Degueldre has been a long-standing proponent of the use of yttria-stabilized cubic zirconia for IMF.

In some cases, a burnable poison (such as erbium, gadolinium, holmium, or boron) or a small amount of fertile additive (such as thorium or uranium) is introduced to improve the neutronic characteristics of the fuel-i.e. by maintaining the reactivity constant over the in-pile irradiation time. Other additives may also be required to stabilize the inert material in the presence of plutonium. IMF pellet fabrication, modeling, and in-pile irradiation tests and studies have been conducted around the world in Canadian, Dutch, French, Japanese, Organization for Economic Cooperation and Development (OECD), Russian, and U.S. research reactors.

This diagram shows the three levels for IMF use in light-water reactors considering homogeneous vs. heterogeneous system or loading concept at the fuel, assembly, and core levels. The fuel is either a solid solution ceramic homogeneously doped with plutonium (red) or heterogeneously doped with some uranium (green), or is a composite material with particulates or microspheres (again plutonium-doped red, uranium-doped green) imbedded in inert matrix material. The fuel assemblies themselves may be homogeneous (all fuel rods in a given assembly contain IMF, red) or heterogeneous (red IMF rods distributed among green UO₂ fuel-e.g. the French Advanced Plutonium Assembly (APA) concept). The reactor core may also be loaded homogeneously (with red IMF assemblies), or the UO₂ core may be partially loaded with some IMF assemblies forming a heterogeneous core loading.

The IMF can be loaded into a reactor in a number of ways. At the fuel-pellet level, the fuel can be either homogeneous (100 percent IMF) or heterogeneous (IMF doped with some UO₂). The fuel assemblies themselves may be homogeneous-i.e., all fuel rods in a given assembly contain IMF-or heterogeneous, with the IMF rods distributed among the UO₂ fuel rods.

The reactor core may also be loaded homogeneously with IMF assemblies, or the UO₂ core may be partially loaded with some IMF assemblies forming a heterogeneous core loading as it is in practice for MOX.

The introduction of the IMF rods into a UO₂ fuel assembly is quite complicated because of the large differences in the neutron spectra of the cell types and their interaction with one another. Many detailed modeling and reactor irradiation studies have been carried out to examine the optimum arrangement of IMF fuel assemblies relative to the UO₂ assemblies in the core.

From such studies it seems clear that additional research on IMF will continue to make a major contribution to the development of nuclear power as a safe and reliable source of energy. The research on IMF has great relevance to the broader area of nonproliferation and nuclear disarmament because it could mean the burning of excess plutonium from power plants.

In the future, standard UO₂ fuel assemblies could be placed together with plutonium IMF assemblies in a reactor such that no net plutonium would be generated. Fuel rods could be configured such that the amount of plutonium generated from the fertile ²³⁸U in the standard fuel is equal to the amount burned in the inert matrix.

Specific studies presented at IMF8 and completed in early 2003

With the study, "Thermal conductivity of zirconia based inert matrix fuel: Use and abuse of the formal models for testing new experimental data," Degueldre not only presented and recommended high-quality data, but also emphasized the purpose of understanding safety-relevant data trends such as ZrO₂-IMF thermal conductivity as a function of temperature or dopant fraction, for example.

The experimental thermal conductivity of an inert matrix fuel material based on yttria-stabilized cubic zirconia-Er_xY_yPuzZr_1-x-y-zO_2-(x+y)/2 (x+y = 0.15, z: [0.05-0.15])-has been measured and intensively studied.

The hyperbolic thermal conductivity trend with temperature known for pure zirconia, ZrO₂ (similar to that known for urania, UO₂), is reduced by the presence of isotopes, impurities, dopants, and oxygen vacancies, which act as scattering centers. They contribute to conductivity reduction to a flat plot with temperature for stabilized zirconia.

It has also been experimentally observed that the thermal conductivity of Er_xY_yM_zZr_1-x-y-zO_2-(x+y)/2 (with M = Ce or Pu, z = 0 or ~0.1 and x+y = 0.15) derived from laser flash measurements is rather constant as a function of temperature in the range 300 to 1,000 K. The thermal conductivity was observed to depend on the concentration of dopants such as YO_1.5 and/or Er_O1.5, or PuO₂.

For example, the bulk material conductivity of Er_0.05Y_0.10Pu_0.10Zr_0.75O1.925 is about 2 W m^-1 K^-1. In this study, the thermal conductivity data of both monoclinic and stabilized cubic zirconia-based IMF are tested with model calculations to understand the experimental data in a semiquantitative way. The derived conductivity models were applied for zirconia, accounting for the effects of isotopes, impurities, and dopants.

The model may be only used in a semi-quantitative way and empirical adjusting corrections are needed.

A comparison of bulk thermal conductivities for ZrO₂ (similar data for UO₂) and Er_0.04Y_0.14Pu_0.09Zr_0.73O1.91, Er_0.04Y_0.14Pu_0.08Zr_0.74O1.91 as IMF and Er_0.07Y_0.10Ce_0.15Zr_0.68O_1.93 surrogate IMF (with Ce as an analog for plutonium). Note: These data are used to calculate the IMF pellet temperature during in-pile irradiation.

The experimentally observed thermal conductivity, which is rather constant as a function of temperature, is justified theoretically and verified semi-quantitatively when applying the model. The thermal conductivity was experimentally observed and modeled to depend on the concentration of dopants such as YO1.5 and/or Er_O1.5, CeO₂ (analogous to PuO₂) or PuO₂. The thermal conductivity of stabilized cubic zirconia-based IMF may be improved by using a minimum of trivalent dopants (Er, Y), which are plutonium-loading connected, and by producing material with very low porosity. These data allow the calculation of the IMF pellet central temperature in-pile during irradiation.

Degueldre also reported on the fabrication and irradiation of plutonium-containing inert matrix fuels for the "Once Though Then Out" (OTTO) experiment, a JAERI, Nuclear Research and consultancy Group (NRG), and PSI project. This irradiation test, in the High Flux Reactor at NRG, Petten, Netherlands, started on Oct. 27, 2000, and was completed in late December 2002. Seven plutonium-containing fuel segments were prepared for an irradiation experiment in which both zirconia-based and spinel-based targets were irradiated up to a plutonium burnup of about 200 GW d m^-3, corresponding to a plutonium depletion of 50 to 60 percent.

Two IMF pellets (solid solution and macro-dispersed) are shown at the beginning of life before irradiation in the frame of the OTTO project. The pellet on the left is a representative pellet made of Er_xY_yPuzZr_1-x-y-zO_2-(x+y)/2 material. The pellet on the right is a composite material pellet. Visible microspheres made of Er_xY_yPu_zZr_1-x-y-z O_2-(x+y)/2 are partially popping out of the pellet's white surface, which is made of MgAl₂O₄ spinel.

For the OTTO experiments, two basic types of plutonium-containing pellets were fabricated: composite pellets and homogeneous pellets. The composite pellets contained spinel as an inert matrix; the homogeneous pellets were based on a zirconia matrix. For the composite spinel fuels, both macro- (250 mm inclusions) and micro-dispersed (25 mm inclusions) fuels were fabricated.

Each fuel contained either uranium or erbium dopant and resulted in four spinel fuels and two zirconia-based fuels. A MOX sample was fabricated for reference. In total seven segments were prepared and irradiated for the OTTO project.

The results of the dimensional measurements, density measurements, ceramographies, and x-ray images of the samples at the beginning of life were given. Some preliminary results of the irradiation were also presented. The on-line analysis of the thermocouples showed good agreement with design calculations derived from thermal conductivity data of the tested IMFs. The neutrographies of the segments made during the irradiation showed limited axial swelling (less than 2 percent) of the fuel stacks, except for the microdispersed material. The segment, which contained a microdispersed spinel-based sample, appeared to be damaged. Extensive postirradiation examinations were performed in early 2003 and are still in progress.

Worldwide interest

The central temperature of the IMF during the OTTO irradiation in the High Flux Reactor, Petten. Comparing the micro- and macro-dispersed IMF composite with the MOX reference.

The Degueldre-Yamashita paper shows the range and depth of current IMF research. It says that from 1995 through 2002, seven workshops on the topic were heldÑthree in Switzerland, one in Italy, one in France, one "within a European Community organization,' and one in the Netherlands.

A total of 350 participants were involved. They came from 17 countries: Australia, Belgium, Canada, the Czech Republic, France, Germany, India, Israel, Italy, Japan, the Netherlands, Russia, the Republic of Korea, Sweden, Switzerland, the United Kingdom, and the United States. Three international organizations participated: the OECD, the Commission of the European Communities, and the International Atomic Energy Agency. Fourteen universities from all over the world (including the University of New Mexico), 17 national laboratories (including Los Alamos and Oak Ridge), and five industrial firms were represented.

Neutron radiography of two IMF segments during the OTTO experiment. IMF composition: (Er,Y,Pu,Zr)O_2-x and (Y,Pu,U,)O_2-x. Pellet diameter 8.00 mm, stack length 67.0 and 67.7 mm, density of plutonium fissile at beginning of life: 0.37 and 0.34 g cm^-3, density 5.80 and 6.02 g cm^-3 respectively. This image was obtained after one cycle.

In the last eight years, 86 papers on IMF have been published in the Journal of Nuclear Materials and in Progress in Nuclear Energy. There have been an additional 88 communications published in five internal reports. Samples and data have also been exchanged, and there have been cooperative activities. IMF8 was the first Initiative for Inert Matrix Fuel workshop held outside Europe. The Degueldre-Yamashita paper said it "was the consequence of the intensive activity of Japan in the initiative." Sixty participants attended IMF8; they gave numerous presentations during ten sessions. The proceedings of the workshop are published in the Journal of Nuclear Materials.

For more information on the upcoming IMF9 workshop to be held Sept. 10-12, 2003, at Sellafield, U.K., see http://www.bnfl.com/website_sellafield.nsf/conference_intro.htm. Degueldre urged members of his audience to get involved in IMF, a new area of science. "The door is still open for IMF," he said.

This article was contributed by Claude Degueldre of the Paul Scherrer Institute, David L. Clark of the Seaborg Institute, and Charmian Schaller of Communication Arts and Services (IM-1).

NMT | LANL | DOE
Phone Book | Search | Help/Info

L O S  A L A M O S  N A T I O N A L   L A B O R A T O R Y
Operated by the University of California for the US Department of Energy