Designs for deep-disposal facilities are dependent on the type of waste and the geology. For SNF and HLW, there is a strong emphasis on physical containment through the use of corrosion resistant canisters or, alternatively, through a corrosion allowance approach that, typically, deploys steel or cast iron canisters with thick walls, which are expected to remain leak-tight for the requisite length of time. Below, we provide three examples of these approaches.

The best-known example of the corrosion resistance approach is the Swedish KBS-3 system,35 which has also been adopted by Finland. The lowland geology of these countries is dominated by granite so that the geological environment corresponds to hard rock in low relief terrain. In the KBS-3 concept, PWR or BWR fuel assemblies are placed inside a nodular cast-iron insert that is surrounded by a 50-mm-thick copper shell: the first supplies mechanical strength and the second corrosion resistance (Fig. 18.5). The complete package, over a metre in diameter, more than 4 m long and weighing over 20 tonnes, is lowered into a vertical deposition hole drilled into the floor of a subterranean passage. Between the container and the rock is an annulus of dried, compacted bentonite, a form of clay. More bentonite is placed on top of the container to seal the hole and, when all the deposition holes have been filled, the passage is backfilled with compacted

rock spoil and sealed. On resaturation, the bentonite becomes plastic and increases in volume, creating a swelling pressure by expanding against the rock and the backfill. The intention is that the bentonite should protect the container from water ingress and rock movement so that it remains intact for a very long period. A variant of the system allows the containers to be put into horizontal deposition holes.

Andra, the national waste management organisation of France, has developed a deep-disposal system for a site that is being investigated in the east of the country.36 Here semi-indurated[32] clay is found at depth on the edge of the Paris basin. The site corresponds to the ‘deep sedimentary basin’ class and groundwater flow is extremely low. France has a policy of reprocessing its SNF so that, for the most part, heat-generating waste consists of vitrified fission products held in relatively thin-walled stainless steel containers; these are produced in various
sizes but, in broad terms, are generally a little over 1m long and 0.4 m diameter. For disposal, each of these is to be placed in a carbon steel overpack with a wall thickness of 55 mm, chosen to provide leak-tightness for several thousand years (Fig. 18.6). The external surface of the overpack has ceramic runners that enable it to be slid into a disposal module — a 40 m long, steel-lined horizontal borehole drilled into the side wall of an underground tunnel. Depending on thermal output, each module will contain up to 20 disposal packages with appropriately sized spacers between them so that, after allowing for a 10 m long closure zone, the whole of the module is used.

For the SNF that will not be reprocessed, carbon steel containers are deployed once again. These have a diameter of about 1 m, a length of more than 4 m and a wall thickness of 110 mm designed to provide a lifetime of at least 10 000 years. Depending on heat output, each container can hold up to three fuel assemblies so that the resulting waste packages are considerably larger and heavier than those for vitrified waste. Up to four waste packages can be placed, equally spaced as before, into a 40 m long, steel-lined, horizontal disposal module. In this SNF variant a bentonite buffer separates the steel liner from the surrounding rock.

A salt dome near the village of Gorleben in northern Germany was designated as a preliminary site for radioactive waste disposal as long ago as 1977. Subsequently, two shafts were sunk to 930 and 840 m and an extensive site

18.2 POLLUX® casks used in the German deep deposal system for spent nuclear fuel (Courtesy of Dr Bernhard Droste, Berlin, Germany).

characterisation programme was performed. Political difficulties have prevented the programme from progressing as quickly as was originally envisaged but, nevertheless, the engineering concepts are well defined. PWR and BWR fuel assemblies are to be dismantled and the fuel rods placed inside thick-walled, self — shielded steel containers known as POLLUX® casks (Fig. 18.7). These have a maximum length of 6.5 m, a diameter of 1.65 m and a weight of 65 tonnes. Transported on a railway system, they are to be laid horizontal in 300 m long, closed-end disposal tunnels and then backfilled with crushed salt.3137 A different system is envisaged for vitrified waste from reprocessing.38 Here, the packages themselves, with no additional containment or shielding, are to be lowered into 300 m deep unlined boreholes drilled into the floor of the repository access tunnels. These two concepts make use of three useful properties of rock salt: its

high thermal conductivity, its stability at high temperatures and its ability to creep to close up voids between the emplaced waste, the backfill and the surrounding host rock. These characteristics allow heat-producing waste to be packed more densely than would be possible in another geological context. The dry environment provided by the rock salt produces very low rates of corrosion so that the containers remain leak-tight for very long periods of time. In this respect, POLLUX® casks are not deployed because a corrosion-allowance approach is being used but, rather, because the shielding they provide removes the need for remote handling.

Conventional mines have a mill where the ore is crushed, ground and then leached — typically with sulfuric acid to dissolve the uranium oxides. The solution is then processed to recover the uranium.

Most of the ore is barren rock or other minerals, which remain undissolved in the leaching process. These solids or ‘tailings’ are separated from the uranium — rich solution, usually by allowing them to settle out. The remaining solution is filtered and the uranium is recovered in a resin/polymer ion exchange (IX) or liquid ion exchange (solvent extraction — SX) system. The pregnant liquor from ISL or heap leaching is treated similarly.

Further treatment for IX involves stripping the uranium from the resin/polymer either with a strong acid or chloride solution or with a nitrate solution in a semi­continuous cycle. The pregnant solution produced by the stripping cycle is then precipitated by the addition of ammonia, hydrogen peroxide, caustic soda or caustic magnesia.

SX is a continuous loading/stripping cycle involving the use of an organic liquid (usually a kerosene-based product) to carry the extractant, which removes the uranium from solution. The uranium is then stripped from the loaded organic liquid using ammonia followed by an ammonia precipitation.

The final chemical precipitate is filtered and dried. Peroxide products can be dried at ambient temperatures to produce a product containing about 80% U3O8. Ammonium or sodium diuranate products are dried at high temperatures to convert the product to uranium oxide concentrate — U3O8 — about 85% uranium by mass. This is sometimes referred to as yellowcake, though it is usually khaki.

In the case of carbonate leaching the uranyl carbonate can be precipitated with an alkali, e. g. as sodium or magnesium diuranate.

The product is then packed into 200-litre steel drums, which are sealed for shipment. The U3O8 is only mildly radioactive (the radiation level one metre from a drum of freshly-processed U3O8 is about half that — from cosmic rays — on a commercial jet flight). In ISL mills the process of uranium recovery is very similar, without the need for crushing and grinding.

Studies of the thorium cycle are ongoing in several countries, such as the USA, Russia, China, Canada, Sweden, Norway, Japan, France and, above all, India. The European Union is also active in fostering R&D actions for the thorium cycle. Finally, it must be mentioned that the IAEA (International Atomic Energy Agency) regularly publishes synthesis documents on this topic. Nevertheless, most of these programmes were limited till today to academic studies. The exception is India and we will, therefore, describe the Indian programme in more detail.

India has limited indigenous uranium resources (1% of the world’s uranium resources) and has difficulties in importing uranium (presently) because of political reasons. With about six times more thorium than uranium, India has made utilization of thorium for large-scale energy production a major goal in its nuclear power programme, envisaging a three-stage approach:

1 Water-cooled thermal reactors, namely, Pressurized Heavy Water Reactors (PHWRs), elsewhere known as CANDUs (CANada Deuterium uranium) fuelled by natural uranium and Light Water Reactors (LWRs) of the pressurized (PWR), boiling water (BWR) or VVER types. In this stage plutonium is produced.

2 Fast Neutron Reactors (FNRs) using this plutonium-based fuel to breed U-233 from thorium. The blanket around the core contains uranium as well as thorium, so that further plutonium (ideally plutonium of high fissile quality) is produced as well as U-233.

3 Advanced Heavy Water Reactors (AHWRs) that burn U-233 and plutonium with thorium, getting about 75% of their power from thorium.

Thorium oxide pellets have been irradiated in Indian research reactors and reprocessed via a simplified THOREX process to recover U-233 (see Section 3.3). Recovered U-233 has also been utilized in research reactor programs. India manufactures ThO2 pellets, which are irradiated as stainless steel clad blanket assemblies in a fast breeder test reactor. Some are also irradiated as Zircaloy clad pin assemblies for neutron flux flattening of the initial core during start-up in pressurized heavy water reactors. The Kakrapar-1 and -2 units are loaded with 500 kg of thorium fuel in order to improve their operation when newly started. Kakrapar-1 was the first reactor in the world to use thorium, rather than depleted uranium, to achieve power flattening across the reactor core. In 1995, Kakrapar-1 achieved about 300 days of full power operation and Kakrapar-2 about 100 days using thorium fuel. The use of thorium-based fuel is planned in the Kaiga-1 and -2 and Rajasthan-3 and -4 reactors, which are today in commercial operation. India is currently building a 500 MWe sodium-cooled fast neutron reactor with, according to the Indian three-phase programme, the possibility, later, of using thorium in the blanket to breed U-233.18

A 300 MWe advanced heavy water reactor (AHWR 300) is now undergoing design and development. The driver fuel will be thorium/plutonium oxide and thorium/U-233. The AHWRs will obtain about 75% of their power from thorium.8 Spent fuel will then be reprocessed to recover fissile materials for recycling.

Based on this overview of past and present developments of the thorium fuel cycle, the next part will summarize the main findings and will give an industrial view of the advantages and weaknesses of thorium as a fertile material for nuclear (fission) energy.

Modern Westinghouse type plants generally have four loops but earlier smaller plants were produced with two or three loops. These included many plants produced by companies such as Framatome (now part of Areva NP) and Mitsubishi under licence. These plants have one steam generator and reactor coolant pump per loop.

Two other US companies developed PWR designs: Combustion Engineering (CE — whose design is now owned by Westinghouse) and Babcock and Wilcox (B&W). In each case their designs have two steam generators and four pumps; i. e. the plants have four cold legs and two hot legs. The Combustion Engineering NSSS is shown in Fig. 10.4 and the B&W NSSS in Fig. 10.5.

The Soviet Union developed its own range of PWRs known as VVERs (Veda-Vodyanoi Energetichesky Reaktor). The smaller reactors, VVER 440s, have six loops each with a pump, loop isolation valves and (horizontal) steam generator. The plant can operate with a loop isolated. The larger, more modern VVER 1000 has four loops, still with horizontal steam generators, but dispenses with loop isolation valves.

The moderator

All Magnox reactors were graphite moderated and CO2 cooled. Graphite is a highly stable material under high temperatures, but it was realised in the very early days of nuclear power that radiation damage would cause atoms to occupy interstitial positions building up stresses in the crystalline structure. These stresses could be annealed out, with the release of Wigner energy. It was loss of control of an annealing operation to release Wigner energy that initiated the 1957 Windscale Pile 1 fire. Wigner energy considerations are important in determining the thermal operating design of the reactor to avoid the need for annealing operations and thermal cycling.

Generally speaking, each channel is centred on a stack of graphite bricks. Interstitial channels permitting the introduction of control rods, flux-flattening elements or instrumentation are located around each channel brick, and the core is held together with radial keying of the bricks. The keying is designed such that changes in graphite dimensions under irradiation and heating could be accommodated without undue distortion to the fuel and control channels.

Surrounding the core is a graphite reflector, to reduce neutron leakage from the core.

Although graphite is highly corrosion resistant to CO2 , irradiation of coolant gives rise to free radicals, which are corrosive. Two undesirable effects may come into play: oxidation of graphite leading to loss of strength and moderating capacity, and deposition of carbonaceous material on fuel or boiler surfaces leading to reduction of heat transfer capacity. Avoidance of steel corrosion is also an important consideration in coolant chemistry.

G. ROSSITER, NationalNuclearLaboratory(NNL),UK

Abstract: It is essential to ensure that nuclear fuel operates within its design constraints, and therefore safely. This requires an understanding of the fuel behaviour under irradiation. The behaviour is complicated by the effects of the neutron flux, fissioning and fission product generation, and microstructural changes in the fuel as irradiation proceeds. Although there are many differences between the behaviour of different fuel types, there are also many commonalities. This chapter describes: (i) the important phenomena that occur under irradiation;

(ii) how and why the phenomena are modelled using computer codes; and

(iii) the advantages, limitations and future trends in the use of such codes.

Key words: nuclear fuel behaviour, nuclear fuel performance.

14.1 Introduction

It is essential to ensure that nuclear fuel operates within its design constraints, and therefore safely. This requires an understanding of the fuel behaviour under irradiation. The fuel behaviour includes both physical (in particular thermo­mechanical) and chemical aspects. The behaviour is complicated by the effects of the neutron flux, fissioning and fission product generation, high temperatures and large temperature gradients, all of which produce microstructural changes in the fuel as irradiation proceeds. Although there are many differences between the behaviour of different fuel types, there are also many commonalities.

This chapter describes: (i) the important phenomena that occur under irradiation; (ii) how and why the phenomena are modelled using computer codes; and (iii) the advantages, limitations and future trends in the use of such codes. In the interests of brevity, only the fuel pin (or fuel rod) design concept used in the majority of reactor types (LWR, CANDU, Magnox, AGR and fast reactor systems) is considered. The focus is on LWR fuel, since this is the most prevalent fuel type internationally. Information on the behaviour under irradiation of the coated particle fuel used in HTR systems can be found elsewhere (IAEA, 1997).

As soon as it was realized that pool capacity will not allow the long-term storage of all of the fuel from a reactor and as the final strategy for and destination of spent fuel were not clear, extended storage capacity became necessary. Dry storage systems started to be developed in the late 1970s and early 1980s. At the end of 2010 there has been more than 25 years of favourable experience on dry spent fuel storage technology. Since it began, the dry storage of spent fuel has evolved into a variety of systems.5,6 Examples are concrete canisters, metal casks (i. e. CASTOR casks in Germany), steel-lined concrete containers, concrete CANSTOR modules (in Canada), and vaults in France, Hungary, the UK and Canada. Many countries developed storage casks (under a variety of names for storage canisters or containers) as self-standing units for containment. In parallel to the development of designs for dry storage, research and investigations were carried out to determine the longevity and stability of materials and spent fuel during dry storage. This was motivated licensing of storage systems. Initially licensing periods were short (from 1 year in the Russian Federation, 2 years in Canada, and 20 years in the USA and Spain to 40 years in Germany). As the need for longer — term storage increased and the confidence in materials and designs grew, the licensing periods were extended. The CASTOR storage containers from Germany can now be licensed for 60 years and this is the tendency in other countries; in the USA they can grant a licensing extension for 40 additional years. The technology for manufacturing dry storage system is mature and we will show just a few examples since detailed information can be obtained elsewhere.6,7 The main requirements for dry storage systems are: containment of radioactivity with watertight or airtight barriers, biological shielding and decay heat removal (cooling).

In addition, the longevity and stability of all materials are becoming even more important when new strategies require very long-term storage (100-300 years) of spent fuel.

Several research projects have investigated the long-term integrity of spent fuel in dry storage.6 Potential mechanisms that may affect cladding integrity during dry storage are: [28]

to promote this mechanism and investigations so far have not shown that these conditions can exist.

Nevertheless, as long-term storage of 100 years or even more is becoming a possibility, and the kinetics of some chemical degradation processes may not be linear, investigations continue to confirm the long-term integrity of claddings. In addition, investigations these mechanisms are affected by transport and extreme accident conditions continue.

Plutonium was first created in 1942 as part of the Manhattan Project when Seaborg bombarded uranium-238 with deuterium particles. He was then able to separate microgram quantities using a bismuth phosphate precipitation process. But kilogram quantities were needed to make a bomb and large nuclear reactors were therefore built at Hanford (first operation 1944) in which many tonnes of natural uranium were exposed to low levels of neutron irradiation. The bismuth phosphate precipitation process was scaled up but the operation was not continuous and, given the small amount of plutonium within the fuel, needed to be repeated several times to reach adequate purity. After the war, the Americans turned their efforts towards solvent extraction, till then used as an analytical method for trace compounds. This separation technique enabled homogeneous phase continuous counter-current extraction operations to reach very high plutonium decontamination factors. Several organic solvents were tested and, eventually, tri-butyl-phosphate (TBP) diluted in an alkane was selected for the PUREX process. This was first implemented at an industrial scale (for military purposes) at the Savannah River plant (USA) in 1954 with another plant opening at Hanford the following year when other methods of extraction stopped.

Similar military plutonium production reactors (in which natural uranium is taken to low burn-up levels) and associated reprocessing plants were started up at Windscale (now Sellafield) in the United Kingdom, Marcoule in France and Mayak in the Soviet Union.

Subsequently, reprocessing began to be used for commercial purposes with a focus on recycling and, especially, separation of plutonium for use in fast reactors. In the USA the first commercial reprocessing plant was opened at West Valley (NY), which was operated by Nuclear Fuel Services from 1966 till 1972. The plant (300 t/year capacity) treated 630 t of fuel. Shutdown was due to high retrofit costs associated with changing safety and environmental regulations and construction of the larger Barnwell facility. Construction of the Morris (IL) facility (General Electric) was halted in 1972 and never operated.

Barnwell (SC) sized for 1500 t/year (Allied Chemicals-Gulf Atomics) never received the required authorisation from the American regulatory bodies and in 1977 all commercial reprocessing on US territory was banned, which is not the case elsewhere in the world. Appendix 1 outlines industrial reprocessing activities in the UK, Japan, Russia and France.

The MA content and composition (i. e. the mix of individual MA isotopes and elements) in the fuel are strongly dependent on P&T mode and objective; they are also key criteria for setting up priorities in terms of R&D programs and associated research facilities. Beyond the need of specific facilities for handling highly radioactive and high thermal load fuel, technological breakthroughs will be needed in areas such as fuel handling systems, discharge storage and transport systems prior to industrial upscaling.

The MA content can vary in terms of MA/Pu ratio from ~0.05 or less in the case of homogeneous recycling in FRs with CR ~1 (e. g. the scenario of Fig. 17.4), to MA/Pu —0.1 in the case of TRU minimization as unloaded from LWRs (e. g. the scenario of Fig. 17.6), up to MA/Pu —1 in the case of MA minimization in dedicated reactors (both ADSs or low CR FRs as in the scenario of Fig. 17.5). In the case of heterogeneous recycling where the MAs are placed in the target, the ratio MA/(Pu+MA) —0.2 is envisaged.

Separation of MAs into the constituents is another possibility that needs to be investigated; this could entail separation of one element or more, e. g. Am only or Am+Cm. These selective approaches, in particular the separation of Cm from Am could allow Cm to be sent for decay storage, given that its main isotope, Cm-244, has a half-life of only 18 years. However, the problem of the final recovery and management of the resulting Pu-240 should also be tackled.

19.3.1 Security during the transport of nuclear material

Security provisions for the transport of nuclear material are established through the Convention on the Physical Protection of Nuclear Material (CPPNM) (IAEA, 1980) for those countries that are parties to that Convention. The Convention entered into force on 8 February 1987 and, as of June 2011, 145 states were parties to it.

An amendment to the CPPNM was promulgated in 2005 (IAEA, 2005). This will not enter into force until the 30th day after the date when two thirds of the states party to the Convention have ratified, accepted or approved the amendment. As of June 2011, 47 states had deposited their instruments of ratification.

A guidance document, which is based on the requirements of the CPPNM and its amendment was issued in 2011 (IAEA, 2011). The document recognizes that the ‘challenges associated with protecting nuclear material from unauthorized removal and sabotage during transport are unique compared to when it is held at nuclear facilities, and thus require a dedicated approach’. Nuclear materials are categorized according to the provisions in the CPPNM and then the protection requirements are graded according to that categorization based upon the total nuclear material present in a single conveyance.

The categorization of nuclear material as mandated by the CPPNM is shown in Table 19.1. The recommended security provisions for all nuclear material should follow a graded approach using this categorization. [33]

The following summarizes the recommendations made in the guidance document, Nuclear Security Series No. 13 (IAEA, 2011). In applying these recommendations, the shipper, carrier and receiver work with the involved competent authority (authorities) to ensure that the entity responsible for each action is clearly defined.

Common requirements for the transport of all nuclear material include:

• minimizing the total time in transport

• minimizing the number and duration of transfers

• protecting the material during transport and temporary storage in a manner consistent with the category of that material

• avoiding use of predictable movement schedules

• predetermining the trustworthiness of involved individuals

• limiting advance knowledge to the minimum number of individuals necessary

• using a transport system with physical protection measures appropriate for the threat assessment or design basis threat

• using routes that avoid areas of natural disaster, civil disorder or with a known threat

• minimizing the time that packages and/or conveyances are left unattended For all categories of nuclear material, steps should be taken to:

• ensure the confidentiality of related information based on the need to know

• establish procedures to ensure the security of keys to conveyances and security locks

• ensure security during unexpected stops and during storage incidental to transport

• ensure that the carrier provides the receiver with advance notification

• provide that all transfers between involved entities are accomplished using prior agreements

• carry materials in closed, locked conveyances, compartments or freight containers with exceptions for heavy packages

• use locks and seals on conveyances, compartments and freight containers

• search conveyances to ensure security provisions have not been violated

• provide sufficient guards and/or response forces based on the category of material and the threat posed

• check the integrity of the packages, locks and seals by the receiver upon arrival of the shipment

For categories I and II nuclear material, additional requirements should be applied, including:

• placing the cargo under surveillance by guards

• ensuring that the receiver confirms readiness to accept delivery

• developing a transport security plan by the shipper and providing the plan to the competent authority for approval

• ensuring that the carrier verifies all physical protection measures are in place in accordance with the plan prior to commencing transport

• using armed guards or compensating measures when justified by the threat assessment

• ensuring sufficient delay is provided by the physical protection measures used

• searching the conveyance prior to loading and then placing the conveyance in a secure area or keeping it under guard

• providing written instructions to personnel with physical protection responsibilities

• providing enhanced protection of confidentiality of information

• providing continuous, two-way communications

• providing an adequately sized response force

Finally, for category I nuclear material, additional requirements should be applied, including:

• providing an enhanced procedure for approval of security plans

• providing competent authorization to ship just prior to commencing transport considering the most current threat assessment

• ensuring guards are appropriately equipped and trained

• providing enhanced security for large packages transported in open vehicles

• establishing and operating a transport control centre (TCC), including providing for secure and continuous two-way communication between involved personnel

• ensuring guards and/or conveyance crew report frequently to the TCC

• applying mode-specific requirements as specified in Nuclear Security Series No. 13 (IAEA, 2011)