Safety in numbers

In 1980 British Nuclear Fuels proposed an investment programme for the Sellafield reprocessing site equivalent to spending £1m a day for 10 years. The investment was needed to enable the oxide fuel from AGR (advanced gas-cooled reactor) power stations to be reprocessed, to update the site’s aged plants and extend the clean-up facilities to protect the environment.

The new plants were designed and built throughout the 1980s and early 1990s. The site is home to the Thorp and Magnox reprocessing plants, the Sellafield mixed oxide plant and a range of waste management and effluent treatment facilities.

Typically, the Thorp (thermal oxide reprocessing plant) cost about £2.7bn and went into operation in 1995.

The plants now have 10 years or more of operational experience, giving an opportunity to appraise the philosophies used in their design.

Design guidance

In preparation for the investment programme, BNFL reorganised its design offices and ran a recruitment campaign, which introduced fresh blood to the nuclear industry. The new recruits challenged the recommendations of the guidance in AECP 1054 (Atomic Energy Code of Practice 1054 – Ventilation of Radiation Areas), first published in 1979.

The nuclear industry Containment and Ventilation Treatment Working Party then set out to develop new design philosophies, and as a result AECP 1054 was reissued in 1989.

The table (below right) compares the recommended air change rates in the two versions of AECP 1054. It can be seen readily that the recommendations for the various areas of a nuclear plant were reduced significantly in the later edition. These reductions accompanied a revised approach to building layout, recognising that nuclear containment involves the interaction of the plant barriers and the ventilation system.

The effects of cutting the air change rate at a large nuclear plant include:

• reduced running costs for fans, as less air moves through the plant
• reduced capital cost of ventilation equipment, as it is smaller
• reduced cost of building space for plant rooms, service voids, etc
• reduced costs of heating the building from smaller volumes and less ventilation
• reduced radioactive waste, with fewer extract systems producing fewer spent filters
• reduced potential radiological exposure to operators/maintenance staff
• reduced aerial effluent discharges as there is less air extracted
• reduced indirect discharges (see later).

Assessing the impact

To assess the potential impact of these changes, Thorp will be used as a basis of comparison. This spent fuel reprocessing plant contains areas of high, medium and low potential hazard, each of which has recommended air change rates as given in the table. To simplify the comparison, Thorp will be assumed to be a low potential throughout, thus the assessment will be conservative as the lowest recommendation is for low potential.

Thorp has an air throughput of 1.5 million m3/h and an average air change rate for the whole building envelope of 1.5/h. For low potential areas, the 1979 version recommended five air changes. Thus, a change rate of 1.5 conservatively puts the systems in Thorp at one-third the size recommended. Fan laws dictate that power consumption is directly proportional to the flow for a given head, thus Thorp running costs, in terms of electricity, are one third of what they would have been.

Based on electricity costs of 5p/kWh (BERR industrial CHP rate), the 1979 recommendations would put the fan energy costs for Thorp at £132m over 25 years. The actual costs on the same basis are estimated to be £46m, a saving of nearly £90m over the plant’s design life.

Ventilation systems in the nuclear industry include HEPA (high-efficiency particulate air) filtration, with the higher potential hazard systems having multiple stages in series. The disposal of spent HEPA filters is a significant issue. The reduction in air throughput brings about a significant drop in this type of solid waste.

Thorp has 446 manual change HEPA filters and 178 remote change HEPA filters, which are the filters required on extract systems for high potential hazard areas. Assuming all manual change filters are disposed of as low-level waste (LLW) and the remote change filters are disposed of as intermediate level waste (ILW), the cost of disposal can be arrived at. A design assumption was that all primary filters would be changed annually and secondary filters changed every two years.

Nuclear waste is extremely expensive to dispose of, with LLW disposal costs being £1500/m3. ILW requires interim storage, prior to the availability of ultimate disposal in a deep repository. The Committee on Radioactive Waste Management (CoRWM) has stated that the repository could be available in 2040, but that all storage facilities are to be designed for a minimum of 100 years’ life. ILW disposal costs before the CoRWM report were put at £120,000/m3. The AECP 1054 (1979) recommendations produce a figure of £1560m for the disposal costs of Thorp’s spent filters, whereas the actual installation equates to £390m.

Another important consideration is the environment and indirect discharges from plants. The government has committed the UK to reducing CO2 emissions, of which buildings currently contribute 46%. The target is a 20% reduction on 1990 levels, which equates roughly to 100 million tonnes of CO2.

Under the 1979 recommendations the annual electricity consumption of Thorp would have been about 12.8MW and is actually 4.6MW. Calculations put the CO2 saving from reduced fan energy consumption at 0.7 million tonnes over 25 years. But then Thorp is not a low potential plant.

The table shows that AECP 1054 (1979) recommends air changes of up to 30 for areas of high potential and Thorp has many of these. Thus, this comparison is conservative in the extreme and the actual savings are much higher. In addition, the smaller amount of waste requires reduced storage facilities and there are similar savings regarding those plants.

Thorp was not the only plant built on Sellafield site during the 1980s and 90s. The list is lengthy and the savings, whatever you choose to measure, are vast. AECP 1054 is guidance across the industry, not just for BNFL, and there have been further savings from plants built on sites elsewhere.

Safety test

The changes to the design philosophies for nuclear ventilation systems were not made to produce savings. The aim was to increase the safety of the plants for the operators, the public and the environment, and the savings were no more than an incidental bonus. Validation of the improvement in safety on the new generation of plants can be obtained by once again considering Thorp.

Early in 2005, British Nuclear Group (as the company was then named) announced there had been a spillage of material in Thorp. The incident was classified as level 3 on the International Nuclear Event Scale (Chernobyl was level 7). Note that British Nuclear Group “announced” the incident – no one outside the plant had been aware of a problem until then.

Dissolved fuel rod liquor had spilled onto the floor inside one of the containments on the plant. It is estimated that about 80m3 leaked from a fractured pipe over a period of time. There was no airborne or liquid release from the cell to the occupied areas of the plant.

The leak of liquor increased the airborne inventory of the affected cell considerably, but the depression regime imposed by the design of the ventilation system prevented loss to the manned areas of the plant. Prevention of the spread of material is the primary function of the containment and ventilation system and the Thorp systems worked extremely well.

The increased airborne inventory would have been drawn into the cell extract system and challenged the two stages of remotely changed HEPA filters. There was no increase in the recorded discharges to the atmosphere from the plant.

Thorp was designed to operate safely under normal and incident conditions, as all modern nuclear plants are. The correct operation of the safety systems designed into the facility to provide containment prevented the liquor spillage from contaminating other areas of the building. The loss of material was limited to the area it occurred in. The material that became airborne was captured by the extract system and successfully removed from the air streams discharged from the plant to the atmosphere.

Conclusion

The successful performance of the containment and ventilation systems in Thorp is validation of the philosophies behind the design. The new approach to nuclear ventilation system design, introduced in the 1980s, has been shown dramatically to have saved energy consumption, reduced waste production and produced nationally significant reductions in CO2 production. n