Ceramic Materials for Long-Term Sequestration of Radioactive Waste

Aaron Palke
March 17, 2011

Submitted as coursework for Physics 241, Stanford University, Winter 2011


While nuclear fission reactors can potentially be safe sources of abundant energy for thousands of years to come a significant unresolved obstacle involves the long-term sequestration of the radioactive nuclides produced. Such radionuclides include fission products such as Cs-137, Sr-90, Tc-99, and I-129 and actinides including U-235, Np-237, Pu-239, and Am-241. [1,2] U-235 and Pu-239 may or may not be removed from the waste material by reprocessing and combined in mixed-oxide fuels. Currently the nuclear industry in the United States employs a once-through fuel cycle in which fuel enriched in U-235 is used once and all of the radioactive nuclides (including fissile U-235 and Pu-239) are considered waste and disposed of. However, if fission reactors are to maintain an important role in the world's energy production profile one can expect reprocessing of spent nuclear fuel and possibly fast breeder reactors (necessarily requiring fuel reprocessing) to become a major part of the nuclear fuel cycle. In this case the sequestration problem becomes one of dealing with so-called high-level waste (HLW) consisting of spent nuclear fuel with the volatile components (i.e. I-129) and economic components (U-235 and Pu-239) removed. The most important sources of radiation after a time period of a few years will be Cs-137 and Sr-90 and isotopes of the actinide elements U, Np, Pu, Am, and Cm not removed by reprocessing. [3,4] The two most widely considered options for disposal of HLW are immobilization in either borosilicate glass or crystalline ceramic materials.

Borosilicate Glass

Sequestration in borosilicate glass material is currently the method of choice in all efforts to immobilize HLW. The advantages of this approach are numerous, but perhaps the most attractive aspects of this method are that borosilicate glasses can accept a wide range of components and that the ease of the scaling-up of this method to a large-scale industrial process. One further positive feature of this method is that borosilicate glasses are immune to amorphization due to radiation damage from decaying radionuclides as these materials are, by nature, amorphous. As of the year 2006 there was a reported 9000 metric tonnes of waste glass in a total of 16842 canisters produced at six vitrification plants in the USA, the UK, France, Belgium and Japan. [2] The main drawback of using glass for nuclear waste is that glasses are intrinsically metastable materials prone to dissolution and devitrification (crystallization).

While safeguards are in place to prevent contact of the glass with groundwater in proposed geological repositories, it is likely that such contact is inevitable in the long-term. [2,5] However, while initial experimental rates of dissolution are unacceptably high, there is a swift reduction to 10-4 of the initial rate due to the formation of an amorphous silica gel phase on the surface of the glass. [5] While there is a solid case for immobilization of radionuclides in borosilicate glasses based on extensive experimental work most of what is known is dependent on the carefully selected environmental boundary conditions of the experiment. The suitability of these boundary conditions to describe a nuclear waste repository which must sequester nuclear waste over a period of 104-106 years is, in some cases, unknown. Therefore, it is prudent to fully explore the other candidates for nuclear waste disposal, the main contender being crystalline ceramic materials.

Crystalline Nuclear Waste Forms

Researchers involved with crystalline nuclear waste ceramics take their cue from naturally occurring minerals known to be highly resistant to dissolution and degradation including pyrochlore [(Ca,REE)Ti2O7, REE = rare earth], zirconolite [CaZrTi2O7], apatite [Ca5(PO4)3(F,OH,Cl)], zircon [ZrSiO4], monazite [CePO4], etc. and which are known to naturally incorporate actinides and natural analogs of fission products [1]. Such a material has never been formed at an industrial-scale but there have been many formulations of what such a material would look like. One of the most successful idea has been SYNROC B formed from a mixture of powdered oxides including SiO2, TiO2, ZrO2, Al2O3, CaO, BaO, SrO, Na2O, and K2O combined with around 10-20 wt% radwaste. The mixture is then pressed and heated up to high temperature (~1300 °C) to promote solid-state recrystallization. The phases present in SYNROC B are a hollandite-like phase (natural hollandite = BaMn8O16) taking the formula BaAl2Ti6O16, perovskite [CaTiO3], and zirconolite [CaZrTi2O7]. In brief, 'hollandite' primarily incorporates Cs from the radwaste while perovskite is responsible for Sr uptake and the zirconolite phase takes in Sr, REE, Cm, Am, Pu, Zr, U, Ru, and many others. [6] These phases together are capable of incorporating most of the radionuclides in HLW.

Of course, SYNROC B is not without its flaws. It has been found that pollucite [(Cs,Na)16Al16Si32O96 nH2O], a natural zeolite mineral, is more resistant to leaching of Cs than 'hollandite' and is thus a more likely host-phase for radioactive Cs. [7] This presents a problem as SYNROC B is specifically formulated to avoid the silicate minerals found in SYNROC A which are more susceptible to leaching of radwaste. [6] Presumably this phase would have to be synthesized separately from the other SYNROC B components. In addition, none of the phases in any SYNROC formulation are capable of taking in 129I which is presumably removed from the radwaste prior to the formation of the crystalline ceramics. The proposed host phase for 129I is thus a variation on the mineral apatite with lead and vanadium replacing calcium and phosphorus respectively [Pb10(VO4)4.8(PO4)1.2I2]. [8] This vanadium-iodine 'apatite' would likely need to be synthesized in an additional step.

The SYNROC family of materials is not the only proposed radwaste disposal system. In some cases almost all of the radionulcides in HLW could be incorporated into a single phase such as monazite [CePO4] which has the added advantage of being nearly immune to amorphization-inducing radiation damage. [9,10] Indeed, the main attraction of crystalline ceramic materials for nuclear waste disposal is their high chemical durability in the presence of water. However, the possibility of radiation damage (primarily due to α-decay) is more severe for these materials than for borosilicate glasses. In fact, natural zircon grains containing a high initial concentration of a UO2 component are known to have undergone amorphization (complete loss of crystalline periodicity) on a geological timescale in a process called metamictization leading to increased solubility. [11] Such a phenomenon leads to complications in U/Pb dating of metamict zircon grains due to increased leaching of U and Pb from the crystal lattice. However, many materials (such as monazite) are immune to such damage due to rapid annealing of the crystal lattice after an α-decay event, and in particular many materials anneal this damage more efficiently at higher temperatures. [9,10] As mentioned above, sequestration of radwaste has not been attempted at an industrial-size scale. One of the main criticisms lies in the relative difficulty and cost of this process as compared to vitrification. However, this process remains a viable option and one that must be considered in a proposal for radwaste sequestration over a geological timescale

© Aaron Palke. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.


[1] R. C. Ewing, "Nuclear Waste Forms for Actinides," Proc. Natl. Acad. Sci. 96, 3432 (1999).

[2] B. Grambow, "Nuclear Waste Glasses - How Durable?" Elements 2, 357 (2006).

[3] A. G. Croff, M. S. Liberman, and G. W. Morrison, "Graphical and Tabular Summaries of Decay Characteristics for Once-Through PWR, LMFBR, and FFTF Fuel Cycle Materials," Oak Ridge National Laboratory, ORNL/TM-8061, January 1982.

[4] W. J. Weber et al., "Radiation Effects in Crystalline Ceramics for the Immobilization of High-Level Nuclear Waste and Plutonium," J. Mat. Res. 13, 1434 (1998).

[5] E. Vernaz et al., "Present Understanding of R7T7 Glass Alteration Kinetics and Their Impact on Long-Term Behavior Modeling," J. Nucl. Mat. 298, 27 (2001).

[6] A. E. Ringwood et al., "Immobilisation of High Level Nuclear Reactor Wastes in SYNROC," Nature 278, 219 (1979).

[7] G. D. Gatta et al., "On the Crystal Structure and Crystal Chemistry of Pollucite, (Cs,Na)16Al16Si32O96·nH2O: A Natural Microporous Material of Interest in Nuclear Technology," Am. Mineralogist 94, 1560 (2009).

[8] J.-M. Gras et al., "Perspectives on the Closed Fuel Cycle Implications for High-Level Waste Matrices," J. Nucl. Mat. 362, 383 (2007).

[9] A. Meldrum et al., "Radiation Damage in Zircon and Monazite," Geochim. Cosmochim. Acta 62, 2509 (1998).

[10] A. Meldrum, L. A. Boatner, and R. C. Ewing, "A Comparison of Radiation Effects in Crystalline ABO4-type Phosphates and Silicates," Mineralogical Magazine 64, 185 (2000).

[11] E. Balan et al., "Metamictization and Chemical Durability of Detrital Zircon," Am. Mineralogist 86, 1025 (2001)