Safe management of the UK separated plutonium inventory: a challenge of materials degradation

The UK holds the largest inventory, worldwide, of separated plutonium under civil safeguards. Here, the importance of materials degradation in managing this inventory to a safe and secure end point is reviewed, together with recent developments, in the context of storage, reuse and immobilisation and disposal.

After more than 50 years of successful operations, reprocessing of spent nuclear fuel will shortly come to an end on the Sellafield site 1 . In so doing, the focus of the Sellafield site mission will shift exclusively to decommissioning of its nuclear facilities. Reprocessing of spent nuclear fuels has afforded a UK inventory of plutonium forecast to be 140 tons at the end of reprocessing operations (of which 23 tons are foreign-owned) 2 , Fig. 1. As discussed here, the continued safe and secure management of the UK plutonium inventory is underpinned by the need to understand the isotopic, chemical and physical degradation of this material, and its impact on the integrity of storage, potential for reuse, and immobilisation for disposal.
The original driver for the UK's inventory of civil separated plutonium was to fuel a fleet of commercial fast reactors 2 . However, although the fast reactor development programme was closed in 1994, reprocessing of nuclear fuel continued, affording the current inventory. In contrast to the spent fuel arising from the UK's advanced gas-cooled reactors, spent fuel from Magnox stations was designed to be reprocessed and was not intended to be directly disposed in significant quantity, due to its reactivity (magnesium alloy clad uranium metal) 2 . Consequently, Magnox derived plutonium constitutes the dominant fraction of the material within the UK plutonium inventory 2 .
The current UK government policy for management of its plutonium inventory was set out in 2011 3,4 , and specifies an intention for reuse of this material in MOX fuel (a Mixed OXide of uranium and plutonium): The UK Government has concluded that for nuclear security reasons the preferred policy for managing the vast majority of UK civil separated plutonium is reuse and it, therefore, should be converted to MOX fuel for use in civil nuclear reactors. Any remaining plutonium whose condition is such that it cannot be converted into MOX will be immobilised and treated as waste for disposal.
This policy was informed by consideration of the credible options for plutonium management, published by the Nuclear Decommissioning Authority, in 2010 5 .
In the international context, there exist considerable declared stockpiles of separated plutonium in France, Russia, the United States of America, China, and Japan, see Fig. 2 and Box 1. Reuse of plutonium as MOX fuel in commercial light water reactors (LWRs) has been achieved in France and Japan, whereas, in contrast, the US MOX Fuel Fabrication Plant was terminated prior to completion of construction, as discussed below 2 . Reuse of civil plutonium as MOX or other fuel in sodium cooled fast breeder reactors has proven more challenging, although Russia and China have ambitious plans for multi-recycling of plutonium in such reactors.

REUSE IN LIGHT WATER REACTORS
In principle, the UK plutonium inventory could be sufficient to fuel three 1100 MWe PWR (Pressurised Water Reactor) or 1600 MWe EPR (European Pressurised water Reactor) units, over a 60 year life time, depending on the core loading. The challenge for the implementation of this policy is both economic and technical. At least 40 reactors have operated with a partial MOX core in Europe and MOX fuel manufacture has been developed at commercial scale by several vendors, most notably Orano (previously Areva), which has produced in excess of 2500 tons; geological disposal of spent MOX fuels is planned within several European programmes 2 . On the other hand, the UK's own experience with MOX fuel manufacture did not achieve the design throughput, and construction of the US MOX Fuel Fabrication Facility was terminated for technical, commercial and financial reasons 2 .
To increase confidence in reuse of plutonium in MOX fuels, research is underway to demonstrate the manufacture of MOX fuel pellets from UK plutonium, which has unique powder characteristics, isotopic composition, and significant americium-241 ingrowth, from decay of plutonium-241 during storage. It is expected that the blending of plutonium batches will enable the americium-241 ingrowth to be adequately managed with respect to MOX fuel manufacture and utilisation 1,2 . Detailed neutronics studies have provided confidence that MOX fuels fabricated from this feedstock will perform acceptably in LWRs 6,7 . However, the manufacture of MOX fuels from UK plutonium remains to be demonstrated convincingly at the required commercial scale. Moreover, no UK reactor operator has yet signalled interest in MOX offtake, and uranium supply is expected to be sufficient to meet projected demand for the foreseeable future 1,8 . Collectively, these factors result in an undeniably weak economic driver for plutonium reuse as MOX fuel in LWRs, although it is a technically plausible solution. Other reuse options, such as in the CANDU EC6 reactor or GE PRISM fast reactors, were determined to present greater technical and implementation risks than reuse in LWRs; nevertheless, the CANDU EC6 system is considered to be a credible option 1 .

OPTIONS FOR IMMOBILISATION
In supporting the UK government in progressing a final decision on plutonium disposition, the Nuclear Decommissioning Authority has commissioned a substantial research programme to determine the quantity of plutonium unsuitable for reuse in MOX fuel and to develop the technology for immobilisation 1 . Given the diversity and characteristics of UK plutonium, which span contaminated residues to fuel quality material, at least three approaches to immobilisation and disposal are under consideration. One method is to immobilise the separated plutonium and residue material in titanate ceramics and glass-ceramics, and recent investigation using plutonium and surrogate species has developed confidence in this approach [9][10][11][12][13][14] . These wasteforms target zirconolite, prototypically CaZrTi 2 O 7 , as the plutonium host phase, which is known to have excellent aqueous durability and radiation tolerance 15,16 . It has been demonstrated that waste packages could be effectively manufactured by hot isostatic pressing, with the advantage of batchwise processing in a hermetically sealed container, produced to near net shape specification [10][11][12][13][14] . Recent research has focused on the development of a furnace containment system to enable the hot isostatic pressing of small scale waste packages incorporating UK plutonium, building on laboratory demonstration studies 17 . Alternatively, a variant of the MOX fuel fabrication process could  Japan. The Tokai Reprocessing Plant produced a mixture of uranium and plutonium oxides from recycle of UO 2 and MOX fuel; this process will also be applied at the Rokkasho Nuclear Fuel Reprocessing Facility. In response to the Great East Japan earthquake, tsunami, and Fukushima Daiici accident in 2011, the Tokai facility was closed and the Rokkasho reprocessing and J-MOX fuel fabrication facilities were subject to additional safety measures which delayed commissioning. The J-MOX fuel fabrication facility will supply MOX fuel for LWRs, which have already utilised MOX fuel produced in France under commercial reprocessing contracts. The future strategy is for plutonium reuse in fast reactors, however, the 280 MWe Monju prototype fast reactor was closed in 2016. The current credible options for management of the UK plutonium inventory are summarised in the following paragraphs, with a focus on the important role of materials degradation in enabling progression to a safe and secure final end point. be utilised to immobilise plutonium in such a ceramic material or in "disposal MOX", sometimes referred to as "low specification MOX", which would be disposed without irradiation 18 .
Whilst there is confidence in the technical feasibility of immobilisation approaches, considerable research and development remains to be undertaken, before such technology could be deployed, which would be "first of a kind" at industrial scale. For example, the wasteform formulation must be adequately underpinned by extensive surrogate and plutonium active studies, at laboratory and demonstration scale, to understand the phase diagram, define the operational envelope and recovery from maloperations. The post-closure safety assessment for the disposal of such wasteforms is complex, due to the coupled nature of the degradation processes, which will progress over 10 5 years 15,16 . For example, self-radiation damage will induce a crystalline to amorphous phase transition during the period of container integrity. This may result in micro-cracking of the wasteform, increasing the surface area available for dissolution reactions, which could plausibly result in differential release of fissile material and neutron poisons incorporated within the wasteform as a safeguard against criticality. Understanding of the impact of these coupled degradation processes on the long term wasteform evolution and alteration, remains fragmented, but is clearly of crucial importance for post closure safety and criticality assessments. In this context, a review and further investigation of relevant natural analogue systems will provide useful long term insight into wasteform degradation mechanisms, at realistic rates, in addition to accelerated laboratory studies. To ensure that the necessary holistic understanding and the evidence base is developed, a roadmap has been developed to plan and guide the UK research programme.

INTERIM SURFACE STORAGE
Both plutonium reuse or immobilisation options are subject to commercial and technical uncertainty and would require at least 15 years to implement, and a further 30-50 years of mission operation. The separated plutonium will, therefore, require several decades of continued storage, irrespective of the final decision to reuse or immobilise. The Nuclear Decommissioning Authority is investing in the design, construction and operation of new fit for purpose stores and plutonium treatment and repackaging facilities, with a lifetime extending to 2120, at a cost of £3.5 billion 1 .
It is known that package degradation during storage depends on complex internal radiation chemistry, understanding of which is, therefore, essential to underpin prolonged storage. In the case of PuO 2 arising from Magnox reprocessing, the package comprises a welded outer steel container, and a polyethylene bagged inner screw-top aluminium container. Initially, these cans undergo depressurisation, as a result of complex thermal and radiolytic oxidation reactions of the polyethylene bag, involving nitrogen oxides formed by radiolysis of N 2 and O 2 19 . PuO 2 produced from ThORP (the Thermal Oxide Reprocessing Plant) was packaged under Ar, in a stainless steel fabricated triple package comprising a screw top inner, vented intermediate, and welded outer container. These packages potentially develop higher internal pressure due to the greater thermal output of ThORP plutonium and outgas of He produced by alpha decay 19 . There remains ongoing debate over the role of PuO 2 reaction with adventitious water within sealed storage containers, producing PuO 2+x and H 2 20 . In the UK context it has been shown that if there is such a direct reaction between PuO 2 and adsorbed water, then surface recombination mechanisms must consume H 2 , inhibiting package pressurisation at storage relevant humidity 19 . At a microscopic level, He accumulation is reported to lead to embrittlement and disintegration of PuO 2 ceramics, after several decades of storage, which may be an important consideration for MOX fuel fabrication and immobilisation options 21 . A small quantity of UK plutonium was packaged in polyvinyl chloride (PVC) bags, thermal and radiolytic degradation of which produced HCl, leading to chloride contamination of the contained PuO 2 22,23 . Recent research has shown that such material may be stabilised for future storage by thermal treatment to remove chloride contamination, followed by repackaging under dry argon 22 . A zirconolite glass-ceramic wasteform has been developed for the immobilisation of contaminated plutonium and residues, for which chlorine solubility in the glass phase has been demonstrated to exceed the conservatively estimated inventory at the envisaged incorporation rate 24 . An upstream heat treatment facility to remove chloride contamination, as specified in the current conceptual flowsheet, would therefore not be required from the perspective of wasteform compatibility, which could de-risk the waste treatment technology roadmap.

THE WAY FORWARD
It may be appreciated that whilst the UK government has a clear plutonium management policy, a final decision on the implementation of reuse or immobilisation is not immediately required. Both an immobilised product and used MOX fuel will require a geological disposal facility as an end point, for which the siting process may require a decade or more. As highlighted here, a considerable programme of research, in partnership with universities and commercial organisations, is underway to inform strategy and a decision on implementation. Whether or not the UK plutonium inventory is reused as MOX fuel, a proportion of the material is known to be in an unsuitable condition for fuel fabrication and will likely require immobilisation and disposal, along with plutonium residues 1,2 . The immobilisation matrix and manufacturing process required to treat these materials would also be applicable to inventory immobilisation, albeit at a larger scale to achieve the necessary throughput. For the foreseeable future, developing the technical maturity of these approaches and understanding the evolution of plutonium during storage are clearly of high importance, to ensure effective waste management, to underpin potential inventory immobilisation, and to assure safe storage. At the same time, it is evident that the UK has committed to a strategy for plutonium management which will continue for many decades. A further priority issue, therefore, is maintaining the knowledge, skills and capability to work with plutonium and highly active alpha materials to support the plutonium management mission 10,22,23,25 . Recognising this challenge, the Nuclear Decommissioning Authority Alpha Resilience Capability programme will seek to ensure this critical expertise is maintained.
The challenge posed by the UK plutonium inventory is complex, neither reuse nor immobilisation are immediately deliverable options, but considerable research has been undertaken, and will continue, to support a final decision on the disposition of this material. As highlighted here, understanding the degradation of plutonium during storage, its impact on MOX fuel and wasteform manufacture, and the degradation of these products in the disposal environment, are of critical importance in assuring a safe and secure end point for this enigmatic material.

DATA AVAILABILITY
There are no primary data to be made available in connection with this analysis. Data relevant to Fig. 2 is available from IAEA: https://www.iaea.org/publications/ documents/infcircs/communication-received-certain-member-states-concerningtheir-policies-regarding-management-plutonium.