Materials scientists face the challenge of devising a passive waste form for long-term immobilisation and disposal of the UK’s surplus separated plutonium. Daniel Reid and Neil Hyatt from the University of Sheffield, UK, outline the possibilities.
One legacy of civilian and military nuclear programmes is a stockpile of separated plutonium (Pu) separated by reprocessing of irradiated nuclear fuel. According to the most recent inventory published by the Health and Safety Executive, the total UK stockpile of separated un-irradiated plutonium was 107.7tHM as of 31 December 2009, of which 27.7tHM was the property of overseas customers. The storage of such a substantial quantity of fissile material, albeit under international safeguards, raises legitimate security and proliferation concerns. The Royal Society report on Strategy Options for the UK’s Separated Plutonium, noted that ‘the chance that the stocks of plutonium might, at some stage, be accessed for illicit weapons production, is of extreme concern’.
Plutonium separation in the civil fuel cycle was undertaken to produce mixed uranium-plutonium oxide (MOX) fuel for an envisaged fast breeder reactor programme, which was cancelled in 1994. One option for reducing the plutonium stockpile is to fabricate MOX fuel for existing and future light water reactors. However, there are both proliferation and waste generation concerns associated with burning MOX fuel. Although a net consumption of plutonium is achieved, fresh plutonium is produced from the uranium present. In addition, the irradiated fuel, bearing highly radioactive fission products, will require eventual disposal either directly, as spent fuel, or in a suitably designed waste form. A proportion of the separated plutonium stockpile is also unsuitable for manufacturing MOX fuel due to the in-growth of the neutron-absorbing isotope Am-241 (from the decay of Pu-241) and other chemical contaminants.
Furthermore, of the UK reactor fleet, only Sizewell B would be capable of burning MOX fuel, but is not licensed to do so. At present, there is no UK policy or underpinning scientific case for the management of irradiated MOX fuel. Therefore, regardless of the plutonium burning scenarios, it is likely that a proportion of the separated plutonium stockpile will prove surplus to fuel requirements.
Clearly, there is a compelling safety and security case to immobilise any surplus and waste plutonium in a passively safe matrix that will prevent dispersal in the environment or illicit use. This presents a considerable challenge to materials scientists who must devise a suitable immobilisation matrix, develop a process for waste form manufacture that is compatible with a nuclear environment, and demonstrate the long-term integrity of the product.
Ceramic materials show considerable promise for immobilising plutonium in a tailored phase assemblage of high chemical durability. The major ‘host’ phase must incorporate plutonium, neutron absorbers (as a guard against criticality) and feedstock impurities (such as Americium) in solid solution. This requires understanding of the underlying crystal chemistry so that these elements are targeted at suitable crystallographic sites.
The host phase should also co-exist with suitable ‘buffer’ phases (such as titanium dioxide, zirconium oxide or silicon dioxide), so that the host phase can self-adjust to unexpected variations in feedstock composition or chemistry. Nature provides a useful insight here, since durable uranium- and thorium-bearing minerals provide excellent target phases for developing synthetic ceramic plutonium hosts.
Using this approach, it is possible to compile a toolbox of mineral-ceramic phases that fulfil the principal requirements of a waste form –
Each of the host phases has particular strengths and weaknesses with respect to –
- Durability – the rate of dissolution in aqueous solution.
- Waste loading – to minimise waste volume.
- Chemical flexibility – to incorporate neutron absorber and feedstock impurities.
- Processing compatibility – sintering temperature, pressure, compatibility with the active environment.
- Volume swelling – arising from α-decay and potential impact on aqueous durability.
- Natural analogues – to build confidence with respect to waste form longevity.
The choice of waste form for immobilisation depends on the relative importance of each of these factors. The table provides a simplified evaluation of potential host phases, however, experimental data in this field are not always directly comparable, impeding detailed comparison of material characteristics. For example, the rate of dissolution depends on temperature, solution pH, sample surface area/solution volume ratio and solution flow rate. A key challenge is to compile a database of pertinent information to allow impartial comparison of material characteristics.
Criticality and safety controls
Any process devised to immobilise plutonium in a passive matrix must be criticality-safe, comply with stringent accountancy and safeguard considerations, avoid generating secondary wastes and be compatible with the constraints of a nuclear working environment. The latter may require remote handling.
The batch-wise nature of ceramics processing is compliant with these demands. Criticality and safeguard requirements are satisfied by processing a known sub-critical mass of plutonium at any time, and verifying its incorporation into the product. Co-processing with suitable neutron absorbers, such as gadolinium or hafnium, for example, can also enhance criticality control. Ceramic processing methods offer the advantage of being readily scaleable. Hot isostatic pressing is being developed to immobilise plutonium containing wastes and residues, and this technology is readily transferable to surplus plutonium. In this area, an ongoing materials science challenge is to simplify and optimise processing technologies to meet the requirements of working in a radioactive environment.
The required period of plutonium containment in a disposal environment, ~240,000y, is relatively short in geological timescales. During this period, 99.9% of 239Pu will undergo α-decay, forming 235U, which is also fissile. Each α-decay process results in several thousand atomic displacements, arising from α-collision and nuclear recoil, potentially leading to amorphisation of damaged areas in certain materials.
The overall effect of accumulated radiation damage depends on time and temperature, as a consequence of recrystallisation of damaged areas. A macroscopic consequence of radiation damage is volume swelling and possible micro-cracking of the wasteform, resulting, potentially, in order of magnitude increases in the dissolution rate.
An engineered repository can only impede radio-nuclides diffusion into the wider environment, hence the first, and most important barrier, against radionuclide release is the wasteform itself. Consequently, the effect of radiation damage on wasteform dissolution is of critical importance. Such studies are extremely demanding and require the use of the short lived, and hence highly radioactive, actinides (for example, 238Pu, half life 87.7 years) to accelerate damage.
A useful method of simulating the radiation damage caused by α-decay is in situ ionbeam irradiation coupled with transmission electron microscopy. Such studies provide a useful comparison of the radiation tolerance of potential plutonium host phases, often with surprising results – for example, under 1 MeV Kr+ irradiation the zirconate pyrochlore Gd2Zr2O7 is extremely resistant to amorphisation, whereas the isostructural titanate pyrochlore, Gd2Ti2O7, is readily amorphised.
Natural analogues also have an important role in this context, validating accelerated radiation damage experiments and models to simulate the effects of radiation damage on durability. Importantly, our group has recently used X-ray Absorption Spectroscopy to demonstrate that ion beam irradiation, under certain conditions, accurately reproduces changes in cation speciation observed in metamict minerals subject to natural α-decay. Understanding of the interplay between radiation damage and the consequent effects on wasteform durability are at an early stage. In addition, thermodynamic and kinetic models of wasteform dissolution and the fate of dissolved species require further enlightenment. The development of our understanding in this area presents a particularly testing materials science challenge, requiring demanding experimental studies coupled with advances in radiation damage modelling.
The ultimate disposal of separated plutonium and other nuclear wasters also presents considerable political and social challenges regarding the location and nature of potential disposal sites. Transparent and open communication of the underpinning scientific and engineering case by materials scientists will be crucial.
Dr Neil C Hyatt, Dr Martin Stennett, and Dr Daniel Reid, Immobilisation Science Laboratory, Department of Engineering Materials, University of Sheffield, Sir Robert Hadfield Building, Mappin Street, Sheffield, S1 3JD, UK. Tel: +44 (0) 114 222 5470. Email: email@example.com
Dr Ewan R Maddrell, National Nuclear Laboratory, Sellafield, Seascale, Cumbria, CA20 1PG, UK. Tel: +44 (0)1925 289800