Disposal of Nuclear Waste
Geert-Jan L.M. de Haas, Frodo C. Klaassen and Ronald P.C. Schram
Nuclear Research and consultancy Group (NRG), P.O. Box 25, 1755 ZG Petten, The Netherlands
Discussions about the reduction of greenhouse-gas emissions have revitalized the discussion about the future role of nuclear power production. Uncertainties about the availability of conventional energy resources like oil and gas have also contributed to a renaissance of nuclear energy. Developments in the Middle East and, only recently, the conflict between Russia and Belarus and Ukraine have once again revealed the vulnerability of the industrialized world when it comes to a secure supply of oil and gas. In addition, the rise of new economic powers like China and India has led to increasing demands on the world markets for raw materials including oil and gas. The prospect of rapidly growing energy consumption rates in the next decades has fueled the debate about the role of other energy resources, including nuclear energy, in order to meet increasing demands.
Over the last four decades nuclear power production has been proven to be a reliable technique, nowadays covering about 35% of the energy needs in the European Union and 17% of the needs world-wide. In contrast to oil and gas, a significant part of today's proven uranium resources are located in politically stable countries; about 50% of the uranium is currently produced in Canada (29%) and Australia (21%). These two factors and commitments to comply to the Kyoto protocol to reduce CO2 emissions have resulted in the return of nuclear energy on the political agenda. The governments of France and Finland have already approved expansion of their nuclear infrastructure.
At the same time it is widely recognized that enlargement of the role of nuclear energy is directly related to the issue of nuclear waste. Enlargement of the nuclear energy's share in the world's energy portfolio requires a socially acceptable solution.
The waste consists of exhausted (spent) fuel rods, which remain radiotoxic for about 130,000 years due to the presence of long-lived isotopes of plutonium and minor actinides like neptunium (Np) and americium (Am), formed during reactor operation by neutron capture, and long-lived fission products like Cs-137, Tc-99 and I-129. For a detailed review on the composition of spent fuel the reader is referred to Bruno and Ewing (2006) in last year's issue of Elements on the nuclear fuel cycle.
Over the last decades new concepts and initiatives for a more sustainable nuclear fuel cycle with emphasis on the management of radioactive waste have been launched. Storage of conditioned spent fuel, or high-level waste arising from reprocessing of spent fuel to recover uranium and plutonium, in underground repositories equipped with engineered barriers is one of the scenarios envisaged for future radioactive waste management. In several countries underground research facilities have become available (e.g. Belgium) or are currently under construction (e.g. Sweden).
Partitioning and transmutation (P&T) is another scenario under consideration. This scenario envisages reduction of the long-term environmental impact of radioactive waste by separation (partitioning) of the most radiotoxic or most long-lived components from the waste and re-irradiating them with neutrons, thereby converting the long-lived isotopes into more short-lived or stable isotopes. Realistic P&T scenarios indicate that the reduction in radiotoxicity equivalent to a period of 150,000 year of natural decay can be reached after 500 to 3000 years (Magill et al. 2003).
It is important to realize that the two scenarios do not only differ from a technical point of view. The final disposal route is an example of open cycle: the fuel is utilized once and then discarded as waste. The P&T route, in contrast, acknowledges the energy potential of irradiated fuel and processes are developed to exploit this potential: a closed cycle. The ‘life cycle' of plutonium is a good example of the latter route.
The bulk of the radiotoxic inventory of the waste, 90%, is made up by plutonium isotopes. The present global inventory of plutonium is over 1,400 metric tons next to about 250 metric tons of weapon-grade plutonium (Ewing 2004). Plutonium is therefore an obvious candidate for P&T studies. About 65% of the plutonium consists of fissile Pu-239 (half life 24,100 yr) and Pu-241 (half life 14,4 yr) (Gruppelaar et al. 1998). Part of the reprocessed, fissile plutonium is currently mixed with uranium oxide and used as MOX (Mixed OXide) fuel in some commercial nuclear power plants. During irradiation in existing light water reactors a maximum of about 50% MOX is applicable. During irradiation plutonium is fissioned (burnt) but at the same time new plutonium is generated by neutron capture in the non-fissile U-238. As a consequence, a net reduction of the worldwide plutonium stockpiles along this line is not feasible. New reactor designs are needed in order to increase the maximum loading of MOX.
Alternatively, effective reduction of the plutonium stockpiles may be achieved by using Pu-bearing fuels, which do not produce new waste. Over the last decade research has focussed on the incineration of plutonium - and other actinides as well - in uranium-free matrices, more generally also referred to as IMF, inert matrix fuels. IMF are composed of a fissile-bearing phase, containing plutonium or a minor actinide like americium, embedded in an inert matrix, i.e. a matrix which does not interact with incident neutrons. The matrices may have a ceramic (e.g. Neeft et al. 2003; Schram et al. 2003) or a metallic composition (e.g. Fenandez et al. 2003). The results of recent irradiation experiments have demonstrated the potential of the concept of IMF (minor) actinide incineration (Konings et al. 2000; Schram et al. 2003). In the next sections an outline on the concept of the IMF is presented, illustrated with the results of a successful irradiation experiment in the High Flux Reactor (HFR) in Petten, The Netherlands.
Composition and fabrication of IMF
Important parameters in studies into IMF fabrication are the chemical composition of the matrix and the distribution of the fissile phase in the matrix. The chemical composition of the matrix is confined to materials (elements) that do not show (significant) interaction with neutrons during irradiation as activation of the matrix may lead to generation of new, extra radioactive waste and/or disturb the neutron balance required to generate a sustainable fission reaction. Examples of inert elements are silicon, aluminium, magnesium and zirconium. In addition, the inert matrix material should be resistant against the different types of radiation that may occur, and be able to accommodate fission products, amongst which are krypton and xenon, and, in some cases, large amounts of helium gas. Moreover, the material should have a sufficiently high melting point and thermal conductivity, and suitable elastic constants to provide mechanical stability. Studied matrix materials are MgO, spinel (MgAl2O4), Al2O3, and ZrO2 (Shiratori et al. 1999; Klaassen et al. 2003; Neeft et al. 2003; Schram et al. 2003).
An important issue concerns the dispersion of the fissile material in the matrix material. A distinction can be made between composite and homogeneous types of IMF. The homogeneous type of IMF is a solid solution of the matrix and the fissile phase. A prominent example is the zirconia-based (ZrO2) fuel which forms a solid solution with PuO2-x: (Zr, Pu)O2-x. Composite IMF consists of an inert matrix which contains a dispersion of either micro- (up to several tenths of a micrometer) or macro-sized (several hundred micrometers) particles of the fissile phase. The size of the particles exerts a significant influence on the behaviour of the fuel during, and after, irradiation. The production of fission products, alpha particles, recoil atoms and neutrons has a profound impact on the periphery of the individual particles. In micro-dispersive systems the damage will be more homogeneously distributed over the fuel (Fig. 1). On the other hand, damaged zones will overlap and result in more swelling than in a macro-dispersive IMF as recently observed (Schram et al. 2003). Clearly, excessive swelling is an important, safety-related, issue.
|. . . . .|
|Figure 2a & 2b|
|. . . . .|
|. . . . .|
|. . . . .|
|. . . . .|
|. . . . .|
|. . . . .|
|. . . . .|
Several routes for the fabrication of the fissile-phase bearing phase, and the IMF, are available. An important consideration concerns the amount of radioactive waste and dust generated during the fabrication process. This is especially important from a radiological point of view. Any handling of radioactive material is based upon the ALARA principle: As Low As Reasonably Achievable. Impregnation of porous yttria-stabilized zirconia sol-gel beads with an actinide (Ac)-bearing nitrate solution is a favoured route as the amount of radioactive waste and dust is minimized (Somers and Fernandez 2005). The impregnated beads are subsequently calcined during which the Ac-nitrate is converted into an oxide, forming a (Ac, Zr, Y)2-x solid solution.
During fabrication of the IMF samples (pellets) the fissile-bearing phase and the inert matrix are thoroughly mixed in the desired proportions and pressed to pellets. Pressures applied generally range between 500 and 600 MPa. The pressed pellets are then calcined at temperatures between 1400 and 1700°C, depending on the type of material selected, to achieve their final density. Typically, the calcined pellets have a diameter between 6 and 8 mm. A high density, i.e. between 90 and 95% of the theoretical density, is required to guarantee sufficient high transfer of the heat generated during irradiation. By far most of the ceramic inert matrices have a poor thermal conductivity properties which may result in temperatures well over 1500°C for ZrO2-based IMF. Too high temperatures may result in melting and other unwanted interactions with the cladding material in which IMF pellets are stacked.
The OTTO experiment
The OTTO experiment has been a joint research project between NRG (The Netherlands), PSI (Switzerland) and JAERI (nowadays JAEA, Japan) on the feasibility of plutonium transmutation using inert matrices. OTTO is an acronym for Once Through Then Out. The OTTO concept envisages burning of plutonium in IMF, designed such that the irradiated pellets are suitable for final storage, i.e. without additional post-treatment. For this purpose two geo-chemically stable matrices were selected: zirconia and spinel. The zirconia-based pellets were prepared from crushed (Zr,Y,Pu,U)O2 particles. The spinel-based pellets were prepared by mixing spinel powder with 20 vol.% micro (< 25 um) and macro-sized (200-250 um) spheres of (Zr,Y,Pu,U)O2. Examination of the surface and the interior of the macro-dispersive type spinel made clear that homogeneous distribution of the spheres was only partly achieved (Fig. 2a), despite the fact that the spheres were mixed with a slurry of spinel (Schram et al. 2003). In contrast the micro-dispersive type pellets showed a massive, homogeneous appearance (Fig. 2b).
The samples were irradiated for 548 full power days in the 45 MW HFR in Petten, the Netherlands during which 30-35% of the plutonium was burnt. After irradiation part of the pellets was closely inspected. During this so-called post-irradiation examination (PIE) the physical and chemical properties of the irradiated pellets are recorded in order to assess their performance during irradiation. Non-destructive techniques include visual inspection, gamma spectrometry, X-ray imaging, tomography and fission gas analysis. Destructive PIE focuses on the microscale properties of the pellets using microscopy, SEM and microprobe analysis. It is good to recall that the samples are highly radioactive. The examinations, including the preparations of the samples, are therefore performed in cells shielded with concrete and lead (hot cells). For handling of the samples and the instruments use is made of manipulators.
Upon inspection of the capsules the cladding of capsule with the micro-dispersive type spinel pellets appeared to be damaged (Fig. 3). Failure of the thermocouple during irradiation had already hinted to (mechanical) problems (note that during irradiation the capsules were stacked in a closed containment so that failure did not result in dangerous situation). Failure of the pellets is clearly visible on X-ray images of the capsules (Fig. 4). The zirconia-based pellets and the spinel pellets with the macro-dispersion on the other hand have remained intact.
Microscopy provides a more detailed, external picture of the thermal cracking of the zirconia-based pellets (Fig. 5). These cracks are typically the result of the high thermal gradients in zirconia, which is a poor conductor. The cracking pattern resembles that of irradiated uranium pellets in common light water reactor fuel. Like zirconia uranium has a low thermal conductivity.
The effects of high temperatures in the core and the dissipation of the heat to the margins of the pellets, where temperatures are lower, is also reflected in the distribution of Cs-137 as recorded by gamma spectrometry. By moving the detector stepwise along the cladding an axial profile of the distribution of several fission products in the pellets, like Cs-137 and Ru-106 is obtained. The axial profile of the capsule with the zirconia-based pellets (Fig. 6) clearly reveals enhanced Cs-137 concentrations at the outer margins of the individual pellets as a result of diffusion away from the core. Less volatile fission products like Ru, Zr and Nb do not show such behaviour.
SEM studies revealed more detailed information with regard to the performance and behaviour of the different types of pellets. The irradiated zirconia-based pellets appeared to have retained a very homogeneous, dense structure. Fine-grained metallic fission products like palladium and ruthenium are evenly distributed over the pellet (Fig. 7). In the case of the macro-dispersive type spinel pellets most of the metallic fission products are concentrated inside the particles (Fig. 8). In contrast to the zirconia-based pellets inclusions of fission gasses (Kr, Xe) are clearly visible. Part of the gasses have apparently diffused outwards, the inclusions neatly aligned along the outline of the spheres. In addition an approximately 10 um thick halo around the sphere has formed. In order to obtain more information about the composition of the halo microprobe analyses of a macro-sphere/halo/matrix section were executed. (Fig. 9). Plutonium shows a clear sharp transition, accentuating the outline of the macro-sphere. Apparently, plutonium does not show any redistribution: it remains within the pellet during irradiation. Fission product concentrations, on the other hand, are clearly enhanced in the immediate surroundings of the sphere, i.e. the halo. The fission products are injected into the matrix over a distance of 10 um, which is the typical range of high energy (~70 - 90 MeV) fission products in matrix material.
As mentioned before the basic thought behind the OTTO concept is the transmutation of plutonium using IMF suitable for direct disposal after irradiation. After having assessed the performance of the various types of pellets during irradiation - and the preliminary conclusion that the microdispersive type spinel IMF is less suitable than a macrodispersive type spinel IMF or homogeneous type zirconia-based IMF - the various pellets will therefore be subjected to a leaching test in NRG's hotcell laboratories. Over the last few years a leaching facility has been designed and constructed. In this facility the pellets will be leached for 6 months using distilled water as leachant at a temperature of 90°C. The start of the leaching experiment is foreseen for the end of 2007. After completion the leached pellets will be characterized studied by light and scanning electron microscopy; the leachates will be analysed by ICPMS at PSI (Switzerland). The results of the leaching test are expected to provide crucial information about the feasibility of the OTTO concept and the properties of irradiated IMF in general.
Bruno J. and Ewing R.C, 2006. Spent nuclear fuel. Elements 2, 343-349.
Ewing R.C., 2004. The environmental impact of the nuclear fuel cycle: climate change, nuclear waste and nuclear weapons. Geochim. Cosmochim. Acta, 2004, 68 (Supplement 1), A13.
Fernandez A., Konings R.J.M. and Somers J., 2003. Design and fabrication of specific ceramic-metallic fuels and targets, J. Nucl. Mat. 319, 44-50.
Gruppelaar H., Kloosterman J.L. and Konings R.J.M., 1998. Advanced technologies for the reduction of nuclear waste. Netherlands Energy Research Foundation, ECN report R--98-008.
Klaassen F.C., Bakker K., Schram R.P.C., Klein-Meulenkamp R., Conrad R. Somers, J. and Konings, R.J.M., 2003. Post irradiation examination of irradiated americium oxide and uranium oxide in magnesium aluminate spinel. J. Nucl. Mat. 319, 108-117.
Konings R.J.M., Conrad R., Dassel G., Pijlgroms B.J., Somers J. and Toscano E., 2000. The EFTTRA-T4 experiment on americium transmutation. J. Nucl. Mat. 282, 159-170.
Magill J., Berthou V., Haas D., Galy, J., Schenkel R., Wiese H.-W., Heusener G., Tommasi J. and Youinou G., 2003. Impact limits of partitioning and transmutation scenarios on radiotoxicity of actinides in radioactive waste. Nucl. Energy 42, 1-15.
Neeft E.A.C., Bakker K., Belvroy R.L. Tams W.J., Schram, R.P.C., Conrad R. and van Veen, A., 2003. Mechanical behaviour of macro-dispersed inert matrix fuels. J. Nucl. Mat. 217-225.
Schram R.P.C., Laan R.R. van der, Klaassen F.C., Bakker K., Yamashita T. and Ingold F., 2003. The fabrication and irradiation of plutonium-containing inert matrix fuels for the 'Once Through Then Out' experiment. J. Nucl. Mat. 319, 118-125.
Shiratori T., Yamashita T., Ohmichi T. Yasuda A. and Watarumi K., 1999. Preparation of rock-like oxide fuels for the irradiation test in the Japan Research Reactor No. 3. J. Nucl. Mat. 274, 40-46.
Somers J. and Fernandez, A., 2005. Fabrication routes for yttria-stabilized zirconia suitable for the production of minor actinide transmutation targets. J. Am. Ceram. Soc. 88, 827-832.