Article - Issue 48, September 2011

Lessons from Windscale's Nuclear Legacy

Peter Mann

Download the article (512 KB)

The owner of any new nuclear power station built in the UK will have to produce detailed plans for its eventual closure before construction can begin. Lessons learnt from decommissioning reactors and waste disposal at the current plants will provide valuable information for future developments. Peter Mann, formerly head of site for Babcock at Windscale in Cumbria, describes the work undertaken on the Windscale Piles and the Windscale Advanced Gas-Cooled Reactor.

The decommissioning at Windscale has led to the development of new techniques that are directly applicable to the UK decommissioning programme. As one of the first countries to operate nuclear power stations, the UK can draw on 50 years’ experience in dealing with reactors at the end of their commercial life. Thanks to this history, the UK’s nuclear industry is now one of the leaders in decommissioning redundant nuclear facilities. The industry has already had to develop techniques to deal with three nuclear reactors at Windscale – the Windscale Advanced Gas-Cooled Reactor (WAGR) and Piles 1 and 2. While some of the challenges at Windscale are specific to those reactors, many of the techniques developed to decommission these reactors are relevant to other nuclear power stations.

The expertise gained will support the decommissioning of the Magnox breed of reactors, most of which have now closed. The last four Magnox reactors, at Oldbury and Wylfa, are scheduled for closure by 2014 having been given lifetime extensions.

There are a number of similarities between Magnox and the WAGR reactors. They all have a graphite core, or moderator, and a steel pressure vessel, along with CO2cooling with similar heat exchangers. These reactors were also built in the same era, using similar materials such as asbestos. Babcock managed the site (as UKAEA) until 2008. It was then contracted to provide senior personnel within the site management team at Windscale, working with the Nuclear Decommissioning Authority (NDA) and Sellafield. It is applying this expertise in working with Japan on an ongoing project decommissioning the Tokai gas cooled reactor.

The devastation at Fukushima power station throughout the Spring of 2011 focused the world’s attention on nuclear safety. The problems posed by a burnt out reactor have already been faced in the UK owing to the fire at Windscale in the 1950s and it is too early to say whether there are relevant parallels that can be used for Fukushima.

In the longer term, the experience from Windscale will help in decommissioning new nuclear power stations. In general terms, this includes reassuring the UK public that the industry can manage the legacy issues surrounding nuclear power. Indeed, the WAGR decommissioning was a European demonstration project, managed by the OECD Nuclear Energy Agency, for safe and cost-effective decommissioning of nuclear reactors. More specifically, the decommissioning has provided the industry with valuable experience in areas such as remote handling, and has underlined the importance of establishing ways to get waste out of the facility and what would happen to it after its removal.

Always innovating

Piles 1 and 2 at Windscale were an extraordinary technological achievement for their time. Removing the reactors has also driven significant engineering innovations in nuclear clean-up. Constructed in the late 1940s, the Piles comprise two graphite-moderated, air-cooled reactors and their associated facilities. Each reactor consisted of 2,000 tonnes of graphite blocks in an octagonal stack, 15 m in diameter and 7.5 m long, surrounded by a biological shield of reinforced concrete 2 m thick.

Following a serious fire in 1957, both Windscale reactors were shut down. All of the fuel was removed from Pile 2 while only a very small proportion of fuel remains in the fire-damaged parts of Pile 1. In the 1990s, Babcock, then UKAEA, began to decommission, disassemble and clean up both reactors.

Phase 1 of decommissioning ended in 1999. The work included putting dams in the original water ducts, previously used to transfer fuel from the reactor to the cooling pond, and in the inlet and exhaust air ducts to seal the massive concrete biological shield surrounding the core. This allowed the installation of a ventilation and monitoring system to replace the existing natural circulation which had been difficult to monitor. The clean-up team also installed monitoring systems in and around the core to measure temperatures, levels of radiation and the amount of radioactive material in the core and airflow. Outside the core, the clean-up team removed accumulations of old fuel and isotope cartridges, used to irradiate materials in the reactor, in the water and air ducts.

Clearing the water ducts was particularly challenging: it involved removing fuel debris from the fire as well as sludge submerged in 750 m3of water. To achieve this, Babcock and its contractors adapted technologies developed for the North Sea oil industry. They deployed remotely operated vehicles on existing rails in the ducts, manoeuvring them from a control console. In this way, the decommissioning team could explore the underwater areas where there might have been radioactive material. The operation provided the tools needed to handle and clear the solids and sludge. This approach not only ensured safe operations but was also highly cost-effective.

Completion of Phase 1 of the clean up left the plant in a safe condition until further decommissioning could begin. Subsequent work has focused on developing a remote retrieval and handling system that can remove the main hazardous materials, remains of fuel and isotope cartridges, from Pile 1’s damaged core, while work on Pile 2 has included removing the reactor’s last remaining isotope cartridges.

An important achievement in the work on Pile 1 has involved gaining approval from the nuclear regulators for the safety case to allow the team to look inside the reactor’s fire-affected zone for the first time since the fire. The team used an endoscope to inspect this area and take pictures. The surveys gave the clean-up team a better understanding of the condition of the area. They used the results to design and develop techniques to remove fuel and other radioactive isotopes, and to devise a strategy for waste treatment, storage and disposal.

A schematic representation of Windscale Pile reactors. This shows the reactor building and the chimney with its filter gallery, from which the air was discharged having passed through the reactor. It also shows the air ducts through which air was drawn into the reactor and the water ducts which provided the export route to the adjacent cooling pond for the discharged fuel copyright Babcock

A schematic representation of Windscale Pile reactors. This shows the reactor building and the chimney with its filter gallery, from which the air was discharged having passed through the reactor. It also shows the air ducts through which air was drawn into the reactor and the water ducts which provided the export route to the adjacent cooling pond for the discharged fuel copyright Babcock

Retrieving the waste

The decommissioning engineers developed specialist retrieval equipment to remove the fuel and isotopes from Pile 1. The so-called Fuel Channel Retrieval Tool (FCRT) can remove fuel elements and isotope cartridges remotely and transfer them to a processing facility next to Pile 1, where the debris will be segregated, put into waste liners and the radiation measured for the records.

To achieve this, an access slot will be cut in the top of the reactor, the concrete pile cap, within a containment area. An overhead crane and manipulator gantry will deploy the FCRT. Cutting the access slot – using a combination of standard industrial techniques and remote operations involving core drills and diamond wire equipment – will allow the decommissioning team to dismantle and remove redundant steel structures and components within the void, followed by remotely operated deployment of the FCRT.

A carbon fibre telescopic mast – carbon fibre minimises weight and avoids compromising the strength of structural components – is installed through the access slot so that the FCRT can be positioned anywhere in the discharge void to retrieve fuel and debris from channels within the graphite core. A combination of ‘end effecters’ (grabs, scoops and loosening tools) operated using camera signals and controls sent through the telescopic tubes, will unblock channels, retrieve fissile material and collect the cartridges and debris in containers. A slightly smaller Isotope Channel Retrieval Tool (ICRT) will remove isotopes in the same way.

The retrieved waste will go to the waste separation cell for encapsulation in an epoxy-based polymer to contain the waste for safe disposal. The use of epoxy resin to encapsulate fuels and radioactive waste has been trialled and attracted significant industry interest which could pave the way for its use in future decommissioning projects (see panel Polymeric Encapsulation).

Once the fuel and isotopes have been removed, subsequent stages will include using remotely operated hydraulic grippers to remove the control and shutdown rods, followed by construction of a modular containment structure, gaining access to the core and removal of the graphite blocks. This will be done using a variety of industry-proven mechanical and thermal cutting tools, including hydraulically or electrically driven devices such as shears, diamond wire band saws and grinders, thermal cutting technologies such as plasma arc and oxy-propane torches, and gripping and lifting tools. Ultimately, this will be followed by demolition of the bioshield, once the remaining material and internal surfaces of the reactor have been removed or decontaminated.

Removal of one of four 190 tonne heat exchangers from the Windscale Advanced Gas-Cooled Reactor copyright Sellafield

Removal of one of four 190 tonne heat exchangers from the Windscale Advanced Gas-Cooled Reactor copyright Sellafield

WAGR

The Windscale Advanced Gas-Cooled Reactor (WAGR) was a prototype for the UK’s second generation of nuclear power stations, the AGRs, which followed on from Magnox stations. WAGR started operations in 1962, shutting down in 1981. A forerunner of a family of 14reactors on seven sites in the UK, the WAGR was CO2-cooled, graphite-moderated and fuelled with uranium dioxide in stainless steel cans. The reactor consisted of a graphite moderator, the structural core, 4.6 m diameter and 4.3 m high, housed in a cylindrical reactor vessel with hemispherical ends. The reactor and its heat exchangers sat in a steel containment building that was 40.8 m high and 41.1 m in diameter, the easily recognisable ‘golf ball’ that came to symbolise nuclear power in the UK.

The WAGR decommissioning programme had several objectives. It set out to demonstrate the feasibility of dismantling a nuclear reactor safely and at acceptable cost. The programme was also designed to establish a route and appropriate authorisation procedures for disposing of the radioactive waste. Another important task was to acquire and record the information, data and expertise that would support the design and subsequent decommissioning of nuclear power plants, especially gas-cooled reactors (see Information
for Posterity
).

The programme has been managed through a series of one- to two-year campaigns that commenced in the 1990s. The campaigns followed a ‘top down’ approach, dealing with the various structures and materials that made up the reactor’s core.

These were principally:

  • the hot box – located above the reactor, this structure received the gas coolant and channelled it to four heat exchangers
  • the neutron shield – constructed of graphite and steel, this absorbed radiation to protect workers on the operating floor
  • the graphite core and reflector – 230 tonnes of
    interlocking blocks of graphite that formed a matrix to position the fuel elements
  • the diagrid and tundish – support structures for the
    graphite core, made up of 90 tonnes of steel
  • pressure vessel – a cylindrical vessel, weighing 118 tonnes, with hemispherical ends.

These components sat within the reactor’s pressure vessel which was itself supported inside a concrete bioshield (a massive structure providing structural integrity and personnel protection from radiation). The sequence of campaigns was:

  • install remote dismantling machine
  • remove operational waste from the 253 fuel channels
  • dismantle the hot box
  • remove the six loop tubes
  • dismantle the neutron shield
  • remove the graphite core and steel restraint structure
  • dismantle the thermal shield
  • reduce the size of the lower structures and remove them from within the reactor’s pressure vessel
  • remove the steel pressure vessel which housed the reactor core, and its surrounding insulation
  • remove the two thermal columns and the outer ventilation membrane from within the reactor vault.

This final part of the decommissioning of WAGR was completed in the summer of 2011.

The Remote Dismantling Machine (RDM) constructed for the project has played a vital role in dismantling WAGR. Operated from a purpose-built control room in a building next to WAGR, the RDM consists of an extendable mast supporting a remotely controlled manipulator. Suspended crane rails enable a three-tonne hoist to travel across the reactor vault into the adjacent cells. Working from the control room, the operators can use the hoist to deploy various tools, such as grabs, oxy-propane torches, shears and grinders. A hydraulic manipulator allows these tools to move in all directions and to reach the full depth of the reactor vessel. During the programme, the decommissioning team designed specialist tools so that the RDM could perform specific tasks during the various campaigns.

Lessons learnt

While new nuclear power stations will have decommissioning ‘designed-in’, the work on WAGR has highlighted that it will still be important to take an adaptable approach to decommissioning. Technology, legislation and policy will inevitably change over the operational life of a nuclear reactor. For example, we can expect to see considerable advances in remotely operated vehicles, mobile cranes, modular and mobile plant and equipment, and the ability to decontaminate equipment. After all, there has been plenty of progress in these areas since WAGR was built, allowing the decommissioning teams to work in ways that were not possible when the engineers designed the reactors.

It is also important to consider the costs of the technologies used in decommissioning. On this front, off-the-shelf equipment can often prove economically advantageous. For example, the decommissioning operators used commercial CCTV cameras on WAGR. Unlike radiation-hardened cameras. commercial cameras may not last as long, but they are considerably cheaper and the total cost is lower.

Work on the WAGR also demonstrated the value of being able to adapt existing systems and structures to avoid the need to build new structures for decommissioning. For example, the waste was removed from the reactor through two of the bioshields for the heat exchangers so that the decommissioning team could benefit from the shielding concrete. To achieve this, the team had to raise the heat exchangers 12 m to make space available. The operators used diamond drilling to create the openings into the reactor vault to provide access for the RDM hoist’s transport system.

A further lesson learnt has been in significantly reducing the number of waste packages. Cutting the waste to appropriate sizes and careful packing ensured efficient use of the space in the waste box. It also proved beneficial to take advantage of radioactive decay to allow disposal as low-level waste rather than intermediate-level waste. The self-shielded waste package also hugely simplified handling and on-site storage.

Looking ahead

Having demonstrated that it is possible to manage the decommissioning of the Piles, completing their clean up is now a lower priority for the NDA than other legacy issues at Sellafield. The Piles can now move into a period of surveillance and maintenance, based on the extensive work done on the safety case, allowing the NDA and Sellafield to allocate resources to projects that have higher priority.

The decommissioning of WAGR has successfully met significant challenges. This work puts the UK’s engineering sector in a strong position to contribute to the growing demand for decommissioning the many nuclear power stations that will reach the ends of their working lives over the coming years.

[Top of the page]