Sellafield has for many years been pilloried by politicians and anti-nuclear campaigners as a “dustbin” for the world’s nuclear industry and has long suffered from a widespread belief that it represents the dirty end of the nuclear fuel cycle.

But the site, and its owner BNFL, may be seeing a turn around in their fortunes. Where there’s muck, there’s brass, and having managed nuclear waste for over 40 years, BNFL can boast a depth of experience which could lead to it becoming one of the top companies tackling clean-up and decommissioning work around the world.

BNFL is in the process of addressing historic wastes which have accumulated at Sellafield from the UK’s Magnox programme. The B38 building comprises 22 silos, each one containing up to 610m3 of waste. The silos contain intermediate level waste, mainly Magnox swarf, the magnesium alloy cladding for the Magnox fuel rods, stripped of the spent fuel, but still contaminated with a surface layer of uranium and fission products. Some silos also contain redundant contaminated equipment. All the wastes in the silos are submerged in water and the older wastes in particular have degraded into a sludge so that the surface of the water has a pea soup consistency whilst the bottom is like a clay. In total the B38 building contains 8000m3 of ILW.

In the early 1980s the Nuclear Instalations Inspectorate and BNFL decided to address the B38 wastes. Some of the earlier silos within B38 in particular were deteriorating and it was clear a long term solution was needed. In water Magnox swarf corrodes to magnesium hydroxide releasing hydrogen in the process. The B38 wastes are continually monitored and controlled as the hydrogen can react with uranium, producing uranium hydride, and explode.

After a considerable planning and design phase, during which engineers assessed a variety of drying, compacting and storing processes, BNFL has developed the Sellafield Drypack Plant (SDP). The first stage of the SDP process is to remove the waste from the silos. A remotely operated mechanical grab will transfer the material from the silos to skips. The skips are transferred within bottom opening flasks on a rail network which connects the B38 site with the SDP building. Once in the SDP, the flasks are positioned above gamma gates and the skips lowered into the separation cave via the flask internal hoist.

Within the separation cave the skips are lowered onto a bogie and driven to a tipping station. They are then tipped onto a vibrating screen. The screen removes oversize material such as bottles, tins and pieces of old equipment. Some oversize material can produce uranium hydride in the drying process and is therefore separated and inspected. The waste is kept wet throughout the tipping and segregation process to prevent any risk of fire.

The sludge which passes through the screen is poured into sacrificial cans using a grab. At the top of the can, space is left for the effluent created in keeping the process clean. Four cans are placed on a stillage and heated to 300?C. Increasing the temperature increases the rate at which Magnox swarf corrodes. The heating process therefore results in the release of a large volume of hydrogen and steam. A baffle arrangement in the can’s lid purges the material with air to prevent a build up of hydrogen. The gases, mainly air, with some water vapour and hydrogen, pass through a condenser to remove the water. It is then filtered to remove radioactivity and released to the atmosphere. The condensed liquids are transferred to the Sixep plant and treated.

The waste within the sacrificial cans is not completely dried, as water entering any future repository could cause the waste to expand if totally dry. Instead the control system reduces moisture content to 5%. The control system contains a complex algorithm which is used to assess the completion of the drying process. Four main variables are off-gas steam content, required heater power, elapsed drying time and off-gas moisture content. Drying time varies from 12-30 hours, depending on contents.

During the drying process, the hot air used for this purpose is kept separate from the purged gases coming off the waste material. There are two back-up systems built into the drying process to ensure that if there is any break down no risks occur.

Once the desired dryness is achieved the drums are crushed using the compactor supplied by the Dutch firm Fontijne. Before delivery to Sellafield, BNFL is carrying out nine months of tests at the factory in Holland, mainly to consider the problems of remotely maintaining and cleaning the equipment. For example every part of the bogie on which the barrels are placed in the compactor has to be remotely replaceable. The compactor produces a 2000 tonne force on the top of the can, reducing it to roughly one third of its original size. Using knowledge of the waste’s origin, combined with measurements of the puck’s weight and size it is possible to determine the uranium and plutonium content. Following drying and compaction the waste takes on a marble quality; it has effectively been converted from a liquid to a solid waste, a much safer and stable form for long-term disposal.

The pucks are left to cool to below 80?C. The control system selects pucks which are then placed inside a drum designed for final disposal. The aim is to maximise the volume available within the outer drums whilst not exceeding constraints set by Nirex for concentration of fissile material. The drums are then transferred to the Waste Encapsulation Plant where they are filled with grout and sealed. The drums will be stored above ground and monitored until a repository is built for final disposal.

Much of the oversize material is also crushed having been cut-up and reduced in size by operators working remotely. Pucks containing oversize and ‘hot’ undersize material have to cool for a certain period.

The SDP is currently being built and is due to start commissioning in mid-1999, reaching full active operation in 2003. To meet its throughput requirements it must process 15 drums a day. At peak production it can manage 21 in a day. It should take 11 years to process all the waste currently in the B38 silos, producing 30,000 drums. However this is a third of the number of drums that would be necessary if the waste were not compacted and will considerably lower lifetime costs. The lifetime of the plant is likely to be 30 years, so there is considerable spare capacity built in which is available for treating other Sellafield wastes.

BNFL’s engineering specialists have carried out the design and project management. With space at the Sellafield site limited, considerable effort was expended at the design stage using 3-D CAD computer modelling. The computer aided design programmes allow two people to do what it would have taken ten to do in the past. It also means many problems which may only have become apparent during the construction, leading to hold-ups and cost increases, are discovered and solved during the design. Safety criteria such as earthquake resistance up to 0.25 G are also integrated into the design phase.

The experience BNFL is gaining in addressing its historic waste legacy, means it is in a unique position to exploit the growing market for waste management and decommissioning, likely to be worth hundreds of billions of dollars. The recent US$6.9 billion contract won with partners Morrison Knudsen for remediation work at the Hanford site in Washington state, USA, as well as the work the company is involved in at Chernobyl, reflect a shifting in the company’s perspective.

“Being an owner operator for more than fifty years, we have developed unrivalled skills in decommissioning our own plants,” says BNFL Engineering marketing director Peter Harrison, “and we can now provide effective services for other clients. Both the West and the East have large decommissioning programmes and with our skills we would expect to achieve considerable inroads into these markets.” Whilst the Western decommissioning market is financed, the money needed to clean up sites in eastern Europe and the former Soviet Union, is not so easily available. Facilitating funding for remediation work, most likely from western governments, is likely to be as much of a challenge as actually doing it.

By going through the difficulties and pressures that BNFL has done over the past twenty years, the company has addressed many problems. In the process it may well have developed processes, in particular the SDP, which can be adapted to other problems arising from the development of the nuclear industry in other parts of the world.