Windscale – core decommissioning goes active

At the end of October 1999, reactor parts were removed from inside the core of the Windscale advanced gas-cooled reactor, WAGR, for the first time. Less than two weeks later, a first waste box was filled with the core material and placed in an intermediate level waste (ILW) storage facility located on site.

This first box contained only sections of neutron shield plugs that had been stored in the reactor’s empty fuel channels. Besides these plugs, the channels also hold bits of fuel assembly and control rods. These components will be removed using the remote dismantling machine (RDM) and an independent lifting/transfer arrangement, which takes the pieces through the waste route that had been constructed within the WAGR plant.

Eventually, the WAGR Project will send some 144 boxes filled with ILW to the storage facility; slightly fewer boxes of low level waste (LLW) will be sent to the nearby Drigg repository operated by BNFL.

On the present schedule, the core and the vessel should be completely removed by 2006. In all about 1200 t of material will be removed. This will leave the biological shield and the existing outer containment in place. The cost of this dismantling operation is about £80 million.

THE WAGR PROJECT

The WAGR demonstration project was set up to prove that UK power reactors could be decommissioned safely, cost effectively and with minimum risk to the environment.

WAGR was the prototype for the second generation gas-cooled reactors (the AGRs) that were developed and constructed in the UK. This 33 MWe plant, which is located on BNFL’s Sellafield site, was shut down in 1981 and its owner, the UK Atomic Energy Authority, undertook to decommission it.

The project started out very much as a research and development effort. According to Terry Benest, head of the UKAEA’s Windscale management team, the object of the exercise was to obtain useful information rather than complete the decommissioning.

The decommissioning programme got underway in a more planned and commercially minded way in the 1990s as the role of UKAEA, which had been responsible for developing nuclear technology since the 1950s, changed to what could be described as a site management agency. The job of UKAEA is now to care for and clean-up old and redundant nuclear facilities and to release them for other uses.

WAGR is now a serious budgetted and targeted project. Its purpose is still to provide assurance that a full-size reactor can be decommissioned cost-effectively, but it is now commercially-driven with as many contracts as possible going out to rigorous competitive tender. The approach is also to adapt current technologies rather than develop new systems specific to the tasks as happened in the past.

In August 1994 UKAEA appointed AEA Technology as managing agent for the project. The prime contractor for the internals dismantling operations is BNFL Magnox Generation. There is also a large number of work packages which are being carried out by various contractors. While the UKAEA expects to gain substantial cost reductions in this way, the contractors expect to gain valuable experience useful for competing in the growing global waste management market.

The project is jointly funded by the UK government and BNFL Magnox Generation* with a contribution from the European Commission.

The dismantling project has been split into 10 campaigns, each requiring separate safety clearance. The work has been divided into a number of contracts which are put out to tender. Besides awarding contracts to carry out the operations, most campaigns also involve an associated tooling and methodology (T&M) effort which is also contracted out.

DECOMMISSIONING STAGES

The first operation undertaken after shutdown was defuelling. This was followed by removal of the top of the reactor and then of the steam generators, which have been sent to the Drigg LLW repository.

This allowed the preparation to be made for dismantling the reactor core. A major part of this job is to plan the campaigns very carefully to ensure that the boxes, which are expensive to dispose of, are economically packed.

ACTIVE COMMISSIONING

The term “active commissioning” is misleading, admitted Jim Craik of BNFL Magnox Generation who was head of this project. As used at WAGR, the term refers to bringing together all the various elements of the core removal operation for the first time.

The removal operation involves:

• Dismantling the reactor using cutting tools deployed by the remotely operated manipulator.

• Routing waste to the sentencing cell.

• Lowering waste to the upper loading cell for monitoring and assay.

• Loading waste into boxes which are transferred to the encapsulation cell, grouted, concreted and dispatched either for disposal at the Drigg repository or for storage at the on-site facility in the case of intermediate level waste.

The ILW store is designed for 50 years of operation. If a deep repository is not then available, the life of the store could be extended; in any case, the waste is recoverable if necessary.

Once the core and PV removal is completed, the project team will review the situation as the next step, removing the biological shield, would produce a very large volume of low level waste. It will be necessary to consider whether to cut it up now or let it decay further. According to Craik, however, it is very unlikely that much further work will be done before 2020 due to the lack of storage and of a long term deep underground disposal facility for ILW.

PREPARING FOR ACTIVE COMMISSIONING

Constructing the waste route

Active commissioning required designing, constructing and proving a waste route which, as much as possible, provides a “production line” for removing the waste. This work highlighted a lack of good records of plant designs. To cope with this, the tooling and handling equipment used had to be designed with quite a bit of flexibility which involved extra costs.

The removal of the steam generators allowed room for constructing the waste route. Additional cells were also constructed within the concrete bioshields of the heat exchangers and an encapsulation plant at below ground level.

The refuelling machine was taken out and a temporary floor, radiation shield and remote dismantling machine installed.

The heart of the dismantling process is the RDM which is in essence a very simple machine. The design and manufacture of this £ 8 million machine began in 1986 while the project was still an R&D exercise and it has only been in position since 1994. There have been some technical problems with software controls which have been overcome.

The RDM is a remotely operated robotic arm on an extendible steel mast, which passes through a rotating shield floor, with a reach of 2.5 m and handling capacity of 35 kg.

A manipulator is attached at the bottom of the mast. Attachments include an oxy-propane torch for cutting, shears and grinders. The box on top has all the needed services for the manipulator – electrical, air, hydraulic, cutting gases, or anything else that is required. The machine is capable of rotating through a full 360 degrees; the mast is offset allowing access to the periphery of the core. There is also a pulley system on a runway beam which also rotates – separately from the RDM – to do the heavy lifting; the mast and manipulator are really only intended for deploying the tools, not for lifting the heavy components of the core structure. The hoist arrangement also transports the components into the sentencing cell, where it is lowered to be assayed, then placed in a waste box.

The boxes, which are made of precast concrete, are unique to WAGR. Once the boxes are filled with waste, a cement grout is added; the box is then capped by casting in situ a reinforced concrete lid.

By today’s standards, this is a very simple solution. The project makes use of off-the-shelf robotic equipment. They see no point in developing it themselves.

The dismantling process

As mentioned above, the reactor is dismantled in a number of stages known as campaigns. An initial campaign with manual intervention has already been carried out.

The current campaign (number 2) takes out the material previously stored in the fuel channels. When completed early this year, this campaign should require nine waste boxes.

Craik described the rest of the process.

Reactor dismantling is essentially from the top down. The hot box is designed to collect the hot gas and feed it to the boilers through coaxial pipes, is at the top.

Removing the hot box is challenging because it requires the development of some difficult cutting processes. It is difficult because of the way the hot box is built: it is a relatively thin walled steel cylinder that has insulation on the inside that is tricky to handle. It has flanged pipes top and bottom which means strip-like cuts through it are not possible which makes it challenging to do with remote machinery. The operator must use plasma arc techniques to cut the tubes from the inside and the main cylinder from the outside.

Having removed the hot box, the team progressively works through the core.

After lifting out the neutron shield, the closed fuel test loops in the reactor must be removed. These highly active systems are dealt with by first filling with concrete grout and progressively lifting and shearing them into sizes that can be managed and boxed. Once all the neutron shield blocks are removed, the main graphite core will be accessible.

The graphite core represents by far the largest volume of material, but does not need much in the way of cutting. The blocks which are keyed together can be cut by a range of tools. The difficulty arises from the assorted pipe work and instrumentation in them, which when lifted out reveal lengths of cable that have to be cut.

Having got the main core out, the steel structure which supports it must be cut out.

The thermal shield is the next area. It is steel bolted and after many years of operation the chance of unbolting it is very remote and will not even be tried. It will be cut up using oxy-propane torches and packaged.

This leads to the most interesting and difficult part of the exercise, the pressure vessel itself. The vessel is made of 4 inch thick steel and covered with about 10 inches of insulation which is intermeshed with a steel mesh material to support it and finely coated with an aluminium skin.

Much work was done with ENTECH to develop a cutting method to cut through this sandwich in a single pass. It uses an oxy-propane with powder injection technique (see panel, p15).

Having removed the pressure vessel, the rest of the job is pretty easy as the dose levels should then be very low. The dismantling team can then take out the final components, the outer ventilation membrane and thermal column sections.

With the completion of this phase of the project, there remains the biological shield to deal with.

THE FUTURE of DECOMMISSIONING

The WAGR Project is expected to demonstrate that UK power reactors can be dismantled safely and cost effectively over a relatively short period. This is likely to have an impact on decommissioning of the commercial units.

At present, the preferred UK strategy for gas-cooled reactors is the “safe store” approach for which the final stage of full dismantling of the plant is left for up to 135 years from shutdown. This time allows activity levels to reduce to where most operations can be done manually. However, critics and regulators will continue to press for quicker decommissioning.

When completed, the WAGR Project is likely to show that it is feasible.