There are a number of nuclear licensed sites in the UK. The NDA owns 20 sites spread across the UK. These sites were formally operated by BNFL and UKAEA, and date back to the start of the UK nuclear programme. There are also further sites operated by other organisations, notably British Energy which operates the Advanced Gas Cooled Reactor (AGR) sites and the Pressurised Water Reactor (PWR) at Sizewell.

The NDA sites contain a wide range of facilities and decommissioning challenges. The sites comprise: [35]

• Twelve reactor sites (five in each of Magnox North and Magnox South, two included within Sellafield)

• A research reactor site at Dounreay

• Two research sites (Harwell and Winfrith)

• A fuel manufacture plant at Springfields

• The Low Level Waste Repository (LLWR).

These facilities will produce the majority of the waste that must be managed in the UK but contributions will also arise from British Energy reactors and others.

The total liabilities associated with these NDA sites are shown in Figure 2.6 Direct decommissioning activities total around £10bn while associated waste management activities make up a proportion of a further £10bn. The total discounted liability for the UK is around £45bn; internationally the challenge is even greater.7

The sites vary markedly in the scale and nature of the decommissioning required. Sellafield is responsible for handling highly radioactive spent fuel and has a correspondingly high decommissioning liability. Some of the plants at Sellafield date back to the early years of the UK nuclear industry and decommissioning these old structures in a safe manner is challenging. Reactor sites, once defueled, have a greatly reduced radioactive burden, which can be further reduced by allowing a ‘‘safe store’’ period of up to 75 years. Reactors therefore form a smaller contribution to the decommissioning liability than Sellafield. Springfields manufactures fuel and does not handle spent, highly active fuel, and so poses less of a decommissioning challenge and requires

correspondingly less money to complete. Research sites pose challenges not present on the other sites, for example Dounreay has stored large quantities of liquid sodium alloy used as a coolant in a test reactor which now must be treated.

The committee on radioactive waste management (CoRWM) has produced an inventory of materials requiring management.8 Their findings are shown in Table 1.

The high level waste (HLW) and recovered uranium and plutonium derive from operations and POCO at Sellafield and so are not associated with decommissioning. The key wastes arising from decommissioning will be intermediate level waste (ILW), low level waste (LLW) and very low level waste (VLLW).

The Sellafield integrated waste strategy9 suggests that around half of the ILW arising at Sellafield will be associated with decommissioning; this is around 140 000 m3, while the Magnox South integrated waste strategy10 suggests about 26 000 m3 of ILW will arise from decommissioning. Further contributions will be made from Dounreay 9000 m3,11 and from Sellafield contaminated land 1600 m3.9 In total at least 60% of the ILW requiring management derives from decommissioning — over 200 000 m3.

Much of the waste arising from decommissioning will be low level waste (LLW) or very low level waste (VLLW). Predicted volume arisings beyond 2030 are dominated by decommissioning activities.12 Total volume arisings are shown in Figure 3 while Figure 4 shows the breakdown into different materials. It is clear from Figure 4 that much of the VLLW material is soil and rubble from decommissioning operations, it would also be expected that a significant proportion of the metals would be associated with vessels and metal reinfor­cement associated with decommissioning. Overall perhaps around 75% of the VLLW (around 1.3 million m3) results from decommissioning operations. A smaller, but still significant proportion of the LLW is also associated with decommissioning.

For many years, protection of the environment from radiation was anthro­pocentric based on the ICRP statement2 that

“The Commission believes that the standard of environmental control needed to protect man to the degree currently thought desirable will ensure that other species are not put at risk. Occasionally, individual members of non-human species might be harmed, but not to the extent of endangering whole species or creating imbalance between species. At the present time, the Commission concerns itself with mankind’s environment only with regard to the transfer of radionuclides through the environment, since this directly affects the radiological protection of man’’.

Thus, the protection criterion for humans (1mSvy-1; see the following chapter by Pentreath) was considered to be sufficiently restrictive that popu­lations of non-humans living in the same environment would be sufficiently protected.

Over the last decade, systems of radiological protection for wildlife have begun to evolve with considerable international and national effort on this issue. In the 2007 Recommendations of the ICRP, the Commission recom­mended the explicit consideration of Radiological Protection of the Environ­ment and recognised the need for advice and guidance, including a clearer framework.3 In 2005, the ICRP formed a fifth committee, which deals speci­fically with the protection of the environment from ionising radiation. Com­mittee 5 proposed a framework for protection of the environment which uses the concept of Reference Animals and Plants (RAPs), designed to be compa­tible with the system of protection used for humans.4 The ICRP also aims to produce a system similar to those used for protection of the environment from other hazards.

The need for a system capable of demonstrating that the environment is adequately protected from the effects of radioactive substances has been recognised by international organisations (e. g. the International Atomic Energy Agency; IAEA)5 and a number of regulators. Environmental protection is now referred to in the International Atomic Energy Agency’s Fundamental Safety Principles.6 The forthcoming revision of the International Basic Safety Standards also mentions radiological protection of the environment. Different approaches have been developed to estimate the exposure of wildlife to ionising radiation. These approaches are used, in some countries, to address require­ments in national legislation to demonstrate that the environment is protected from anthropogenic releases of radioactivity.7 12 Radiation protection has not always been the driver of this process, in some countries a system of protection is required to address conservation legislation.

The ICRP4 has focussed on Reference Animals and Plants (RAPs) which are defined as follows

“A hypothetical entity, with the assumed basic biological characteristics of a particular type of animal or plant, as described to the generality of the taxonomic level of Family, with defined anatomical, physiological, and life — history properties, that can be used for the purposes of relating exposure to dose, and dose to effects, for that type of living organism’’.

The RAPs are hypothetical entities and not intended to represent a particular species. Commonly, in assessment approaches linked to tools, the approach taken to address the wide range of different wildlife species is to use ‘‘reference organisms’’. The selection of reference organisms has considered the need to encompass protected species, and different trophic levels and exposure path­ways.13,14 Reference organisms have tended to be defined at a broad wildlife group level (e. g. soil invertebrate, predatory fish, terrestrial mammal etc.). The definition of reference organisms used in the integrated Environmental Risk from Ionizing Contaminants Assessment and Management (ERICA) approach developed by EC researchers15 is

“a series of entities that provide a basis for the estimation of radiation dose

rate to a range of organisms which are typical, or representative, of a

contaminated environment. These estimates, in turn, would provide a basis

for assessing the likelihood and degree of radiation effects’’.

In contrast, for some approaches specific species rather than wide groupings have been considered.11,16

To assess the risk of radioactivity to wildlife we need an approach which contains the following components:

(i) transfer of radionuclides to wildlife;

(ii) dose conversion coefficients relating internal and media activity concentrations to estimate absorbed dose rates to wildlife; and

(iii) interpretation of the biological effects of radiation to determine the risk to wildlife.

There are currently three comparatively comprehensive assessment models which are freely available: the ERICA Tool which implements the ERICA Integrated Approach,17 RESRAD-BIOTA18 which implements the US Department of Energy’s Graded Approach9 and the England and Wales Environment Agencies R&D128.11,14

CLINT A. SHARRAD, LAURENCE M. HARWOOD AND FRANCIS R. LIVENS*

ABSTRACT

The waste materials generated in the nuclear fuel cycle are very varied, ranging from the tailings arising from mining and processing uranium ore, depleted uranium in a range of chemical forms, to a range of process wastes of differing activities and properties. Indeed, the wastes generated are intimately linked to the options selected in operating the nuclear fuel cycle, most obviously to the management of spent fuel. An open fuel cycle implies the disposal of highly radioactive spent fuel, whereas a closed fuel cycle generates a complex array of waste streams. On the other hand, a closed fuel cycle offers options for waste management, for example reduction in highly active waste volume, decreased radiotoxicity, and removal of fissile material. Many technological options have been pro­posed or explored, and each brings its own particular mix of wastes and environmental challenges.

1 Nuclear Fission as an Energy Resource

The vast majority of nuclear reactors which operate, or have operated, have produced energy from uranium fuel. The neutron-induced fission of a uranium nucleus typically yields around 200 MeV of energy (compare ca. 4 eV per atom in the oxidation of carbon to CO2), so the energy density of nuclear fuel is very high, and the volumes of fuel required and waste produced are relatively small.

* Corresponding author

Issues in Environmental Science and Technology, 32 Nuclear Power and the Environment Edited by R. E. Hester and R. M. Harrison © Royal Society of Chemistry 2011

Published by the Royal Society of Chemistry, www. rsc. org

Nevertheless, because of their radioactivity, wastes from nuclear energy pro­duction are potentially very hazardous over long timescales and their man­agement is often both politically contentious and technically demanding. While there is considerable international cooperation in the area of nuclear energy, through for example the activities of the IAEA or the OECD’s Nuclear Energy Agency, and international commercial activities such as fuel reprocessing and fuel manufacture do occur, the management of radioactive wastes is presently seen as being a national responsibility. The following discussion is largely written from a UK perspective, although it draws on examples from overseas where appropriate.

In 2008, 439 nuclear power plants were in operation in 31 countries accounting for 15% of the world’s electricity production1 and as of January 1999, a total of 2532 nuclear detonations have been performed.2 Many sites involved in the production, reprocessing and storage of nuclear materials have contributed to environmental radionuclide release through discharge, spillage, accidents and testing facilities. A summary of such sites and their associated

contamination issues is given in Table 1. The number of nuclear facilities is expected to increase in the near future as governments renew their interest in nuclear power due its reduced CO2 emissions and the energy security it provides when compared to other forms of electricity generation.

Engineering considerations for multiple barrier structures have been the subject of extensive research, with design considerations varying between countries and according to waste type (see Table 1). For example, the latest Swedish design concept for spent fuel which has also been adopted by Finland (KBS-3 concept; see Table 1) favours waste encapsulation in corrosion-resistant copper canis­ters, with waste packages emplaced in isolated boreholes lined with bentonite, and with clay or bentonite blocks and crushed host rock used a backfill (see Table 1).7,8 In contrast, the Belgian Spent fuel/HLW concept favours waste encapsulation in steel supercontainers, which are axially emplaced in disposal tunnels, with Ordinary Portland Cement (OPC)-based concrete used as a buffer and cement-based materials used as backfill (see Table 1).9,10 Generally, international plans for ILW favour co-disposal alongside spent fuel and HLW, albeit in separate areas of the same repository (see Figure 2 and Table 1). Furthermore, ILW waste conditioning is typically in concrete/grout with encapsulation in steel and emplacement in boreholes or caverns that are backfilled with concrete, grout, clay and/or crushed host rock (see Table 1; review by Hicks 2008; ref. 11).

It is also worth mentioning that the USA currently hosts the world’s only operational GDF, the Waste Isolation Pilot Plant (WIPP). This facility receives transuranic (TRU) wastes containing uranium, plutonium and other actinides, which are broadly similar to UK ILW. The site selection process for this facility began in the 1950s and waste emplacement commenced in 1999. WIPP is housed in stable salt rock formations, and design features include waste packaging in steel containers, waste emplacement in horizontal bore­holes or shafts, and the use of MgO as a backfill material.11 However, whilst WIPP has commenced geodisposal of US defence related wastes, the long term management pathway for US civil wastes is uncertain (and subject to current review) due to the US Department of Energy (DOE) recently with­drawing their license application for a HLW repository at Yucca Mountain, Nevada.12

The dose conversion coefficients can be used to estimate the unweighted absorbed dose rate from media and organism activity concentrations. For internal exposure the following equation can be used:

Dbnt = X Cb * DCCiui (5)

i

Where:

• Dbnt is the absorbed internal dose rate for reference organism b;

• Cb is the average concentration of radionuclide i in reference organism b (Bqkg-1 fw); and

• DCCbnt { is the radionuclide-specific dose conversion coefficient for internal exposure defined as the ratio between the average activity concentration of radionuclide i in the organism b and the dose rate to the organism (pGyh-1 per Bqkg-1 fw).

For external exposure the following equation can be used in terrestrial ecosystems:

D bxt = X vzY, Crzf * DCCbx

z

Where:

• vz is the occupancy factor, the fraction of time that organism b spends at a specified location z in its habitat;

• Crzf is the average concentration of radionuclide i in the reference media of a given location z (Bq kg-1 fw, soil); and

• DCCbbxtzi is the dose conversion factor for external exposure defined as the ratio between the average activity concentration of radionuclide i in the reference media corresponding to the location z and the dose rate to organism b (pGyh-1 per Bq unit media).

External DCCs from beta and alpha emitters are comparatively low and may be assumed to be zero for some radionuclides in some approaches.46

Weighted total dose rates (in pGyh-1) can be calculated as:

DCCint = wf lowp ‘ DCCint, lowP + wf p+g ■ DCCint, p+g + wf a ‘ DCCint, a (7)

DCCext = wf lowp ‘ DCCext, lowp + wf p+g ‘ DCCext, p+g (8)

Where wf are the weighting factors for various components of radiation (low energy p, p + g and a).

Although there is no agreement on wf for wildlife, currently most assessment approaches are using broadly similar values with default radiation weighting factors of 10-20 for alpha radiation and 1-3 for low beta radiation and 1 for beta-gamma radiation.13,14,47 For a-radiation weighting factors, these values are broadly consistent with the upper bound on the range of variation reported by Chambers et al.48 in relation to deterministic endpoints (mainly mortality). Currently, the estimated doses for wildlife do not take account of tissue weighting factors as used in human dosimetry.

The ICRP have derived dose conversion coefficients4 for 75 radionuclides for the Reference Animals and Plants using the methodology of Ulanovsky and Prohl49 developed for the ERICA Tool.13

Vives i Batlle et al.50 presents a comparison of unweighted whole-body dose rates from ten different approaches being used (or developed) to assess the exposure of wildlife to radiation.

The separation of plutonium from irradiated nuclear fuel was originally developed in the nuclear weapons programmes of the 1940s and 1950s. Since nuclear weapons require plutonium of high fissile (239Pu) content, irradiation times in production reactors were short and burnups were low, no more than a few hundred MWd tonne 1. In these feed materials, the fission product loadings were therefore also low, and the content of higher actinides was small, so that they presented a much less severe challenge to separations technology than modern, high burnup fuels from civil reactors. In addition, little attention was paid to waste management in early weapons’ production programmes.

All large scale separations depend on the diverse redox chemistry which characterises the mid-actinides (see Table 3). In the media of interest for separations, the substantial chemical differences (for example in solubility in aqueous or non-aqueous solvents, or in affinity for a complexant) between the linear dioxo “actinyl” ions, MO2+/2+ formed by oxidation states v and vi, and the ‘‘simple’’ M3+/4+ ions formed by lower oxidation states generally provide the basis of useful separations. Solvent extraction processes, which are the mainstay of current technology, often exploit the differing affinity of different actinide species for selective complexants, usually O-donor ligands. The following examples are not an exhaustive description but serve to illustrate the diversity of processes which have been explored or used, and of the waste streams which can be produced.

This technique is based on the in situ delivery of chemical oxidants to the contaminated media to destroy the contaminants by converting them to harmless compounds. Typical oxidants applied in this process include hydrogen peroxide, potassium permanganate, ozone and dissolved oxygen. A common application of this procedure, based on Fenton’s Reagent, involves the addition of hydrogen peroxide and an iron catalyst to the contaminated area, generating a hydroxyl free radical:

H2O2 + Fe2+=> Fe3+ + OH + OH* (1)

This free radical is capable of oxidising complex organic compounds, such as TCE, PCE, dichloroethylene (DCE), benzene, polyaromatic hydrocarbons (PAHs) and polychlorinated biphenyl (PCBs), with any residual hydrogen peroxide decomposing into water in the subsurface. Fenton’s Reagent oxida­tion is most effective in acidic environments (pH 2-4) and becomes ineffective under moderate to strongly alkaline conditions.84 Ozone can oxidise con­taminants directly or through the production of hydroxyl radicals and is also most effective in acidic conditions. Due to its instability and high reactivity, ozone is produced onsite and requires injection via closely spaced delivery points. Permanganate is typically provided as a liquid or as solid potassium permanganate (KMnO4) but is also available in sodium, calcium or magnesium salts. Permanganate reactions occur at a slower rate compared to ozone and peroxide and, depending on the pH, can destroy the contaminant through direct electron transfer or free radical oxidation. An advantage of permanga­nate use includes an operable pH range of 3.5 to 12.84

This method was used in a field study at the A/M Area of the Savannah River Site where undissolved dense non-aqueous phase liquid (DNAPL), including TCE and PCE, contamination was present. The treatment test operated over six days and used hydrogen peroxide and iron sulfate to generate hydroxyl radicals to destroy approximately 600 pounds of DNAPL-contaminated soil in the target area. After the trial period, 94% of the targeted DNAPL was destroyed85 at a total cost of $511k for the project demonstration.

104 Richard Kimber, Francis R. Livens and Jonathan R. Lloyd

Naturally occurring radioactive isotopes are either primordial (e. g. 40K, 238U or 232Th; present from the creation of earth), including their decay products, or cosmogenic (formed by cosmic rays).4,29 Primordial radionuclides and the

ЛОО OOP

decay products of U and Th represent the more significant problem with regard to environmental contamination,30 but concentrations vary significantly in the environment and depend on local geology.4 Migration of and exposure to naturally occurring radionuclides can be significantly enhanced by industrial activities, such as mining and mineral processing, in particular in production of phosphate,31,32 oil production and combustion of coal (which contains trace

72 1 72 72 72 Л

quantities of radionuclide) in power stations. , , The radionuclide concentrations in different types of coals range from 12-435 Bq kg 1 for 238U; 21-309 Bq kg 1 for 226Ra; 7.5-56 Bq kg 1 for 232Th and 6-398 Bq kg 1 for 40K.34 When coal fuel is burnt in power plants the ashes generated are enriched in metals and radionuclides. The amount of ash released into the atmosphere from coal-fired power plants can vary from 10% in an old plant, to 0.5 % in modern emission-controlled power plants.35 In addition, coal burning also releases radon into the atmosphere.

Although globally the release of naturally occurring radionuclides through industrial activities is a relatively minor source of contamination, compared to civil and military nuclear programmes, it can still result in local elevated levels of contamination. Flues et al. (2002)35 found a one — to three-fold increase in the natural radionuclide concentrations (232Th, 226Ra and 210Pb) within a 1 km distance of a 10 MWe coal-fired power plant, and elevated radon levels have been reported in and around coal power plants but the dose is less than the recommended occupational exposure limit.36

1.2 Accidental Release

Accidental release of radionuclides from nuclear facilities or from other sources (e. g. industrial or medical) is a less significant source of radioactive contamina­tion in the environment and the largest releases have been due to the accidents at Chernobyl and Fukushima.4 At Chernobyl, 1.76x 1018 Bq of 131I and 8.5x 10 Bq of 137Cs were released into the atmosphere;4 at Fukushima preliminary calcula­tions estimate that 1.5×1017 Bq of 131I and 1.2×1016 Bq of 137Cs were released between 11th March and 5th April 2011 (ref: Japan Nuclear Safety Commission, press release 12th April 2011). Nuclear accidents are discussed in detail in Chapter 3. From other sources, one of the most serious incidents occurred at Goiania, Brazil, in 1987. A radiation source from a cancer therapy machine was scavenged from an abandoned clinic and sold to scrap dealers, who opened the source, containing 50.9 TBq of 137Cs in the form of a luminescent powder.37,38 The material attracted a great deal of interest and was distributed to family and friends of the scrap dealers. As a result, 249 people were contaminated, with 21 suffering acute radiation sickness or radiodermatitis; four people died and another six were in a serious condition. Seven sites covering 5000 m2 in Goiania were found to be highly contaminated and clean up involved the demolition of houses and the construction of repositories for the waste.38

The process of optimisation of protection is intended for application to those situations that have been deemed to be justified in the first place. The principle, with restriction on the magnitude of individual dose or risk, is central to the system of protection and applies to all three exposure situations. It is defined by the ICRP as the source-related process to keep the likelihood of incurring exposures (where these are not certain to be received), the number of people exposed, and the magnitude of individual doses as low as reasonably achiev­able, taking economic and societal factors into account. This process of opti­misation over several decades has resulted in substantial reductions of occupational and public exposures, and is key to the entire approach currently advocated for radiological protection. Essentially, it is always aimed at achieving the best level of protection under the prevailing circumstances through an ongoing, iterative process that involves:

(i) the evaluation of the exposure situation, including any potential expo­sures (the framing of the process);

(ii) the selection of an appropriate value for the constraint or reference level;

(iii) the identification of the possible protection options;

(iv) the selection of the best option under the prevailing circumstances; and

(v) the implementation of the selected option.

4.1 Dose Limits

Dose limits only apply to planned exposure situations (but obviously not to medical exposures of patients). For occupational exposure in planned exposure situations, the limit is expressed as an effective dose of 20 mSv per year, averaged over defined five year periods (in other words, 100 mSv in five years), with the further provision that the effective dose should not exceed 50 mSv in any single year. For public exposure in planned exposure situations, the limit is expressed as an effective dose of 1 mSv in a year, but in special circumstances a higher value could be allowed in a single year, provided that the average, again over defined five year periods, does not exceed 1 mSv per year.

The limits on effective dose apply to the sum of doses due to both external exposures and to committed doses from internal exposures arising from the intake of radionuclides. Occupational intakes may be averaged over a period of five years to provide some flexibility. Similarly, the averaging of public intakes over a period of five years would be acceptable in those special circumstances where averaging of the dose to members of the public could be allowed.

Dose limits do not apply in emergency exposure situations where an informed, exposed individual is engaged in volunteered life-saving actions, or is attempting to prevent a catastrophic situation. For informed volunteers undertaking urgent rescue operations, the normal dose restriction may be relaxed. However, responders undertaking recovery and restoration operations in a later phase of emergency exposure situations should be considered as occupationally exposed workers, and thus protected according to normal occupational radiological protection standards, and their exposures should not exceed the occupational dose limits. (Female workers who are pregnant, or are nursing an infant, should not be employed as ‘‘first responders” undertaking life-saving or other urgent actions).

Notwithstanding the basic scientific method adopted by the ICRP, its approach to the selection of dose limits necessarily includes societal judgments applied to the many and varied attributes of ‘‘risk’’. Not only would these judgments probably be different from one operational context to another within any given society, but they are also likely to differ from one society to another. Providing general guidance is therefore not that easy, and the ICRP makes it clear that it is for this reason that its guidance is intended to be sufficiently flexible to allow for national or regional variations.