End of life management on the Harwell site

7 February 1999

The research reactors and many other nuclear facilities at the UK Atomic Energy Authority’s Harwell nuclear establishment ceased operating by the early 1990s. The UKAEA then embarked on a unique project – decommissioning one of the world’s major nuclear facilities. To accomplish this, the Authority devised a strategy optimising safety, environmental, technical, staffing and cost factors. All of the major redundant facilities have now been decommissioned to Stage 2 and many others demolished. By 2020 it is expected that the only remaining radioactive facilities at Harwell will be three reactors and an ILW store, all in a passively safe condition. The remainder of the current site will have been redeveloped as the Harwell International Business Centre. by K F LANGLEY & R A SIMPSON

Harwell was established as a centre for atomic energy development in 1946 on the site of a former RAF (Royal Air Force) airfield. Most of the nuclear reactors, laboratories, workshops and other facilities, which were built in the late 1940s and early 1950s, are now redundant. In particular, the material testing reactors, DIDO and PLUTO, were shut down in 1990 together with other supporting facilities. A programme of decommissioning work has been in progress for a number of years with the objective of caring for and, at the appropriate time, safely dismantling the redundant facilities. The ultimate aim is to progressively delicense those areas which are no longer required for nuclear purposes and redevelop the Harwell site as a diversified international business centre.


Decommissioning strategy (see panel opposite) is developed for the Harwell site as a whole within the overall company planning framework. A plan has been derived which balances and optimises these factors. This is achieved by integrating programmes for decommissioning individual facilities and optimising waste management routes within the site plan. The costs and associated safety and environmental risks can be balanced with the safety and environmental benefits from decommissioning and savings in the costs of continued operation, building refurbishment, maintenance and infrastructure. Within this framework, UKAEA also demonstrates that the use of public funds for the company and sites decommissioning programme is optimised by financial appraisal using a 6% discount rate (defined by HM Treasury). Once the long term plan has been optimised, the impact of changes is almost always towards increased overall cost, as is discussed later.

The current status of the Harwell programme is given in the panel on page 24.

The main emphasis for the near term future is on four areas:

• Processing approximately 1000 m3 of intermediate level waste accumulated since the earliest days of nuclear development at Harwell in order to place it in a passively safe state for long term storage in a newly constructed Vault Store (cf para 179 of document Cm 2919: “When the demands of safety are overriding, waste must be treated as necessary to improve storage conditions”). Processing the waste is expected to take about 10-15 years. Long term storage will be required pending the availability of a deep repository for an indefinite period which could be 100 years if necessary.

• Replacement of the existing active liquid effluent treatment plant with a modern, more effective and simpler plant. The system was designed for throughput of 50 000 m3 per year of low level active effluent. As a result of decommissioning activities and rationalisation of the drainage system, arisings are already below 5000 m3 and are expected to fall to 1000 m3 within five years.

• Delicensing and redevelopment of the eastern end of the Harwell site. This involves further decommissioning of several minor facilities, the demolition of two former Aircraft Hangars which house GLEEP and other plant, and careful surveying of the land to demonstrate that no significant hazard from radioactivity remains.

• Remediation of a number of areas both on and off the licensed site with largely chemical, but some radioactive, contamination.

Currently, the B459 PIE building is being kept operational so that it will be available over the next 3-5 years to repackage some intermediate level solid wastes. This will involve removing radioactive materials from old shielded containers, separating the true ILW fraction for storage in the Vault Store and disposing of the remainder of the material as low level waste. When this work has been completed, B459 will be decommissioned to Stage 3.

It is envisaged that by 2020, the only remaining UKAEA redundant radioactive facilities at Harwell will be three reactors (DIDO, PLUTO and BEPO) and the Vault Store in a passively safe state receiving no further wastes. The reactors will remain for a further 20-50 years, by which time it will be feasible to decommission them. Most of the remainder of the site will have been delicensed and redeveloped. The site plans show the current and expected coverage of the UKAEA managed licensed site in 1998, 2010 and 2020. The area between the waste store (B462) and DIDO and PLUTO reactors shown on the final map may also be delicensed subject to commercial interest. Remaining radioactive work on the Harwell site will be carried out by other organisations such as AEA Technology (formerly the commercial arm of UKAEA, now a privatised company).


The key factors in determining past and future success are:

• Decommissioning management.

• Contractorisation versus the use of own resources.

• Contracted out decommissioning.

• Records management.

• End-point issues.

• Understanding infrastructure costs.

• Developing a framework for delicensing.

The following describes some of the main issues and lessons learned:

Decommissioning management

Harwell is licensed under the Nuclear Installations Act by the Health and Safety Executive (HSE). This places a number of licence conditions on the UKAEA as operator of the site. In particular, we must:

• Ensure the safety of the site.

• Operate under a safety management system acceptable to the HSE.

• Remain in control of the plant as operators and exercise close supervision over contractors.

• Carry out decommissioning of redundant plant under plans submitted to the HSE.

UKAEA is also required by the Department of Trade and Industry (DTI) to demonstrate best value for money in the use of public funds. This means justifying expenditure by carrying out financial appraisal of options using the Treasury discount rate and making the maximum use of competitively-tendered, fixed price contracts for implementation work. At first sight these constraints appear to be mutually exclusive. HSE has a preference for decommissioning as early as possible, and for use of experienced, in-house personnel where ever possible. DTI prefers that, provided plant can be maintained in a safe condition, decommissioning should be phased to minimise discounted lifetime costs. In practice, both requirements can be met by developing a holistic approach taking into account infrastructure savings, by judicious planning and by use of both in-house staff and competent contractors.

The UKAEA safety management system for contractors is based on a good specification of the task, establishing the competence for the task of both the contractor and their staff, control of the safety case, endorsement of all method statements, control of staff on the facility and regular, often continuous, UKAEA supervisory presence at the work site.

The commercial perspective is to transfer the task and its project risks to a contractor in return for an agreed payment. However, because of the long history and complexity of the Harwell site no significant task can be undertaken without consideration of nuclear and site interaction issues for both redundant and operating facilities. As the holder of the Nuclear Site Licence, the UKAEA remains absolutely responsible for the safety of operations and activities on the Harwell site. While every effort is made to produce a definitive specification, in each of the major facilities decommissioned there has been a significant uncertainty about the extent of contamination. This aspect is discussed in more detail below.


After closure of any facility, the removal of active materials requires methods that are usually an extension of the normal operational procedures of the plant. This is best handled by the operators. This type of operation has been undertaken successfully at Harwell through the removal of fuel from the reactors, and the removal of experiments and the associated active materials from redundant facilities.

When post operational clean-out (POCO) is complete there is a transition into decommissioning activities which requires careful management. UKAEA at Harwell has implemented two different strategies for this transition:

• Retention of a core team of operators for the decommissioning activity.

Examples of this are Stage 2 decommissioning of the DIDO and PLUTO Reactors and post-irradiation examination (PIE). For DIDO and PLUTO the operators remained UKAEA staff and at the end of the process they were re-deployed. For the PIE facilities, the staff were divested to NNC as part of a decommissioning contract. In both cases, the operators successfully made the transition to a decommissioning role.

• Redeployment of staff to other activities and facility handover to a decommissioning management team.

This approach has been successfully implemented at Harwell for facilities including B351, B47 and B336.28. Each building contained a number of facilities, experimental and prototype plants. After handover, a documentation review was undertaken, followed by a detailed radiological and physical survey to define the state of the facility. Small liabilities and residual waste items were removed by staff on a term contract managed by the UKAEA management team. When it was possible to clearly define either part or all of the remaining liability, major contracts were let to complete the decommissioning.

The success of both approaches is reflected in the fact that the buildings have been decommissioned within budget and at a lower than anticipated dose uptake. It is possible to change from the use of operators to contractors during Stage 1 decommissioning and successfully decommission to a ‘green field’. The experience at Harwell suggests that the key requirement for both approaches is a focused decommissioning management team with a good site/facility understanding and a strong supervision capability. Provided these are met, for all Stage 2, Stage 3 and Remediation decommissioning, contracting can be a successful strategy consistent with the requirements of the nuclear site licence.

Contracted out decommissioning

Decommissioning of B351 was carried out under a series of 17 separately competed contracts involving six main contractors under the direction of an in house management team. Term contractors were also used to deal with a large number of small tasks. The work was completed for £13.5 million, nearly 20% under budget. In contrast, single contractors were used for stage 2 and 3 decommissioning of the B35 facility, the B336.17 LLW incinerator and B336.28, which were much smaller facilities. The first two of these projects went significantly over budget because the contractor found substantially more contamination than had been expected from preliminary surveys. In the case of B336.28, the project was terminated after the building had been demolished because, again, more extensive contamination was found underneath the base slab than expected. In this case, it was decided to carry out final remediation along with the remainder of the liquid effluent treatment plant site, where this facility was located.

In principle, the extent of commercial and technical project risk transferred to the contractor is defined through the contracting process. In practice, the presence of unidentified contamination and structures not defined within the specification limits the scope for transfer of such risk. The management approach must be tailored to the prevailing circumstances. B351 was a large facility containing many redundant prototype and research plants. For the smaller facilities, it appeared possible to scope the extent of decommissioning from building records and recent surveys. The potential for additional contamination and structures was identified as a project risk because of the age and early use of the buildings. In the event, this project risk became reality and resulted in contract variations. However, all of the buildings have been successfully decommissioned.

The most effective strategy has been proven to be to maintain UKAEA management of the overall project until it is possible to contract defined packages of work which can be scoped to present the contractor (and UKAEA) with an understandable and manageable risk. A thorough understanding of the liability to be decommissioned is essential to writing a specification that can be used by both the contractor and UKAEA to understand and scope the risk. UKAEA safety supervision by knowledgeable staff also allows the best solution for both parties to be derived when unexpected liabilities or issues are encountered. In the case of B336.28 the “New Engineering Contract”, with its requirement for a quick response to “Compensation Events” provided a well-defined but flexible framework for managing unexpected finds of contamination.

Infrastructure, the hidden cost

The infrastructure for the Harwell site includes security, safety services, health physics, waste management operations, facility management, projects, site management and company overheads. Decommissioning allows reduction of these costs in stages as areas are decommissioned and released. Incremental expenditure on one facility can lead to enormous savings on site infrastructure costs as a whole. By removing fissile material from the B220 Radiochemical Building, it was possible to reduce the security requirements which resulted in substantial savings. Decommissioning a number of minor facilities has reduced site maintenance costs, facilitated the modernisation of the site utilities infrastructure and is expected to increase income from tenants.

By 2020 the infrastructure will be required only to maintain the residual facilities (DIDO, PLUTO, BEPO and the waste stores) and will be significantly less than current levels. Assessment against the planned decommissioning programme can identify the potential overhead and infrastructure cost reductions at each significant stage. A delay which ripples through to the end of the programme leaves unnecessarily high support costs for every year of the delay. So the aim of achieving good value for money is reconciled with making steady progress in decommissioning the site in accordance with Cm 2919.

Within the net cost of managing, operating and decommissioning the Harwell site of about £22.5 million per annum, the annual overhead and infrastructure cost is about £15 million. A one year delay within the next five years to the Harwell decommissioning programme results in a discounted cost of about £6.5 million to the overall programme. While this is a simplistic model to demonstrate the point, the overall decommissioning programme for Harwell has many dependencies and the programme will be difficult to recover without additional expenditure, the inevitable consequence of delay.


For a Nuclear Licensed Site, the options for any phase of the nuclear mission are:

• To continue with related nuclear activities.

• To retain facilities on the licensed site under care and maintenance where there is a long term justification, such as to allow for the beneficial effects of radioactive decay or to store waste prior to disposal.

• To close down the facilities, remove the liabilities and develop the site for other purposes.

Each of these options is part of the forward plan at the Harwell site over the next 25 years. AEA Technology continues to carry out nuclear business activities from the site, whilst the B462 Complex contains the Harwell Vault Store, and a possible associated Box Store which will remain until the availability of an ILW repository. Current plans are that the DIDO, PLUTO and BEPO reactors remain for some years to benefit from radioactive decay before final decommissioning and demolition. These facilities continue as “prescribed installations” under the Nuclear Installations Act (1965) on a licensed area.

The remainder of the site is either in preparation for, or undergoing decommissioning, or is being used for other business purposes. There are no plans for further use of other facilities for nuclear activities. This planned future development for alternative use is best carried out in a delicensed regime.

The Harwell site has a complex history and has been subject to a diverse range of nuclear operations. Delicensing and re-use of facilities has not been undertaken before in the UK for such a complex site. In the past UKAEA has delicensed a small area containing some office buildings at Harwell and others have delicensed specific site areas and individual facilities. There have been a number of recent legislation and standards issued defining criteria for remediation of both chemical and radioactive contaminated land (see List). Consequently, there is little in the way of precedent or experience to guide the current delicensing process.

In 1997, the UKAEA initiated the project to delicense 25 hectares at the North Eastern end of the Harwell site. To date, the work has focused on the legal, technical and financial issues associated with delicensing. Section (3) of the NIA(65), enables the licensee to end their “Period of Responsibility” and vary the boundary of the nuclear licensed site, by demonstrating to the HSE that “there has ceased to be a danger from ionising radiation from anything on the site or that part thereof”.

The certainties of the 1960s create real problems in the 1990s. Recognising that absolute safety is unobtainable, UKAEA has presented a proposal to interpret ‘no danger’ as ‘tolerable risk’ in line with the documents listed. Consistent with Cm 2919 and HSE consultation UKAEA proposes to adopt a delicensing target criterion of a risk of death of 1x10-6/year, as being representative of ‘tolerable risk’. From Cm 2919 the UKAEA will relate the risk of 10-6/yr to a dose rate of 0.02 mSv/yr. If this level cannot be achieved an assessment will be made, in accordance with the ALARP (As Low As Reasonably Practicable) principle, to determine whether further clean-up is required prior to delicensing.

Site specific radiological modelling is being used to translate the risk target into a radionuclide concentration, against which measurement of residual radioactivity in buildings and on land areas can be compared. This technical assessment will be supported by arguments based on historical records and the decommissioning/remediation undertaken to achieve a clearance state suitable for delicensing.

The current business model for this area of the site shows the considerable benefit that can be obtained from delicensing and the return of the land to more general use. It is currently planned to subdivide the area into a series of zones and delicense to release land value on a phased basis. The advantages of this approach are:

• Delicensing cases will be smaller and easier to prepare, review and approve.

• The delicensing programme is responsive to the decommissioning programme.

• The release of the land can be timed to maximise value according to the local property market.

• Phased realisation of land value can potentially enable the later stages of the project to be self funding.

The current decommissioning programme results in an earliest end point for complete delicensing the North Eastern area of approximately the year 2008.


Successful decommissioning at Harwell over the last eight years provides confidence that complete decommissioning of a nuclear licensed site is feasible in the long term. However, this cannot be fully achieved until a deep repository for intermediate level waste is available. Until then the strategy is to minimise the hazard on the site and the cost of managing it. The management team has demonstrated ability to take the plans forward and the contractor base has been established to support the plans.

UK decommissioning strategy

The UK Government believes that the process of decommissioning nuclear plants should be carried out as soon as it is reasonably practicable to do so (Cm 2919 (1995), Review of Radioactive Waste Management Policy, HMSO, London). In practice, this means carrying out decommissioning in three stages: Stage 1 (sometimes known as Post Operational Clean-Out [POCO]) is normally carried out immediately after closure. It can involve operations such as removal of fuel, coolant and non-fixed items of equipment. This may be followed by a period of care and maintenance pending further decommissioning. Stage 2 involves dismantling and removing most of the remaining nuclear material. Stage 3 is the removal of all the remaining radioactive hazard, with the site returned to unrestricted use. The timing of the Stages of 2 and 3 depends on a number of factors in selecting the optimum decommissioning strategy. These factors include safety of the public and workers, protection of the environment, the availability of waste routes, the time required to develop decommissioning techniques and plan operations, possible benefits from radioactive decay, avoidance of undesirable transmutations (eg decay of Pu 241 to Am 241), the availability of suitable personnel (and other resources) and optimisation of the use of public funds.

Harwell programme status

The table below lists the main nuclear facilities at Harwell and summarises their current status. All Harwell reactors have been decommissioned to Stage 2, although in the case of PLUTO some of the ancillary buildings remain in use. Both DIDO and PLUTO have operating ventilation systems to control humidity and minimise corrosion rather than control radioactive contamination. The reactors are effectively in a passively safe state. The radioactive inventory is mainly Co 60 with a half-life of about five years. By about 2040, radioactive decay will allow the reactors to be dismantled manually and most of the fabric disposed of as low level radioactive waste. BEPO has been decommissioned to Stage 2 and is in Care & Maintenance (C&M) to allow for Co 60 decay until about 2060. In the case of GLEEP, which was a very low energy reactor, Stage 3 decommissioning can be carried out as soon as a route for disposal of the graphite can be agreed and the other uses of the building containing the reactor have come to an end. A number of other facilities have already been decommissioned to Stage 3. These include: B35 (an RAF workshop, previously used for uranium fuel and other fabrication), B451 (a building containing a small swimming-pool reactor known as LIDO), B47 (an RAF workshop converted to a beryllium laboratory and decontamination centre), B351 (the main Chemical Engineering Laboratories), B336.17 (the Harwell LLW incinerator), B336.28 (an effluent treatment pilot plant building) and B540.2 (Variable Energy Cyclotron and remote handling cells). The former sites of these facilities are now either a grass area or a base slab awaiting remediation as part of other site plans. In the B220 Radiochemical Facility, about 120 plutonium-handling glove boxes have been dismantled using robotic equipment. This was more economical than manual methods for the large number of glove boxes involved. Much of this work is described elsewhere (see References). Further decommissioning is underway, including B393.6 (a Post Irradiation Examination (PIE) facility).

List of key legislation and standards associated with clearance of radioactively contaminated land

Radioactive Substances Act, 1963 (Revised 1993). Ionising Radiations Regulations, 1985. HSE Tolerability of Risk, 1988 (Revised 1992). European Council Directive 96/29, Euratom Basic Safety Standards, 1996. International Atomic Energy Authority, Application of Radiological Protection Principles to the Clean-up of Contaminated Sites, 1997. Control and Remediation of Contaminated Land, A Consultation Paper, DETR, 1997. Radiological Protection Objectives for Land Contaminated with Radionuclides, NRPB, 1998. Generalised Derived Limits (Revised), NRPB, 1998.

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.