The Sell-by Date

If the average Magnox power station were a pot of yoghurt, sitting quietly on the supermarket shelf, the local Environmental Health Officer would have been called in long ago. Not only because the pot of yoghurt would be well past its sell-by date but because its container had large cracks from which it had been leaking yoghurt by-products, all over the shelf, for years.

More than that, the Environmental Health Officer would, in all probability, have prosecuted the supermarket for making inspection of the yoghurt so difficult. The pot of yoghurt would have been given a decent burial and all the consumers would have slept happily in their beds.

Alas, the same cannot be said for magnox nuclear power stations as for our hyperthetical pot of yoghurt. Nor indeed for the Nuclear Installations Inspectorate (NII).

The history of the Magnox programme has been littered with system failures, corrosion by carbon dioxide, embrittlement by neutron bombardment and, yes, like our yoghurt pot, large cracks (up to 3 metres in length) in pipework.

They say that those who do not learn from the lessons of history
are doomed to repeat its mistakes.

The Magnox power programme has been littered with expensive mistakes.

The History

1967: Partial meltdown at Chapelcross

Following refuelling, a channel of fuel overheated and melted. Two men had to work in the hot reactor for 25 minutes before the fuel elements could be removed. The reactor was out of service until 1969. Similar partial meltdowns have take place in french reactors of similar design at St Laurent A1 in 1969 (closed for a year) and St Laurent 2 (1980 closed for 2 years).

1968: Corrrosion Leads to Derating

The CO2 coolant gas was officially found to be oxidising (coroding) the inaccessible mild steel componenets in 1968. The reactors were down-rated (derated) to work at a lower capacity; this was in order to reduce the gas outlet temperatures by 40C and thus reduce corosion.

Bradwell was derated in 1971 and 1980, Sizewell A in 1967 and 1971.

Because the components are inaccessible ultrasound scans are used to estimate the corrosion damage. At Bradwell, when these estimates were compared to the real thing the real corrosion was found to be much worse.

1976: Releases of Coolant gases at Wyfa and Dungeness

Information released in response to a Parliamentary question. Both releases were reported under the "Dangerous Occurrences Act"

1978: Cracks and Defective Welds in Gas Ducts

Severe cracking was found in the expansion joint section of the coolant gas ducts at Bradwell and Dungeness A. Both stations were closed for approximately two years. Similar cracks wer found at Sizewell A in 1982 resulting in extensive closure. Hinkley A had cracks of 3 metres long (1980).

1980: Sizewell Thermocouple Shields Corroded.

The coolant gas temperature is measured by thermocouples, which in Magnox are set behind shields. In 1980 oxidation led to cracks in the shields and a five year programme of repairs began. Other stations have undertaken similar programmes.

1981: Earthquake Resistance Questioned.

Calder Hall military reactors were found to be vulnerable to damage at an earthquake force of 0.05g (which occurs once every 100 years). This caused wide concern. The NII held that the Magnox reactors (which can stand 0.15g) were safe as no earthquake had occurred in Britain greater than 0.25g. However the British Geological Survey reported that a 0.4g quake took place in 1580.

The NII funded the British Geological Society's research......

1986: Corrosion in Standpipes

The steel channels in the reactor roof (stand pipes) through which nuclear fuel is loaded and unloaded, were found to be corroded at Hinkley A.

1986: Trawsfynydd Releases 15 Tonnes of Coolant gas.

A faulty valve caused 15 tonnes of radioactively contaminated coolant to be released. A site emergency was called and staff were not permitted to leave the plant. The cloud spread 5-10km from the plant.

1991: Trawsfynydd Shut down Due to Embrittlement.

In 1991 the NII decided not to allow Trawsfynydd to continue to generate because of embrittlement of welds. It never restarted and is currently being decommissioned

EMBRITTLEMENT

In 1988 Bradwell had its long term safety review (LTSR) carried out by the NII. The NII expressed concern about embrittlement of welds and vulnerable gas ducts. Neutrons which, inevirably escape from the reactor core had been found to cause the embrittlement.

Embrittlement is an extraordinary unpredictable condition.It's like putting your car in a garage one night with a small crack in the windscreen and finding that the next morning the whole screen has shattered. Ifbrittle joints have to be cooled rapidly (after a loss of coolant accident for example) there is real danger of rupture. As is the case with Magnox, a problem for one reactor is usually a problem for another.

In the league table of embrittlement Trawsfynydd lead by a long way followed by Hinkley, Sizewell, Dungeness and Bradwell.

Of these five Dungeness and Sizewell are still operating    (Jan 2002).

Many Magnoxes have areas that cannot be physically inspected; inspired guess work (usually called "engineering judgement" by the industry) is used to calculate the rest. In yoghurt-pot-land it is as if the Environmental Health Officer were unable to sample the yoghurt and, despite the bulging container and cracked lid, still gave the supermarket the OK to keep it on the shelves for another 6 years.

So why on earth do the NII allow these reactors to continue running?

Can the Nuclear Industry exert more pressure on the NII than public opinion?

In the USA, in Massachusetts, a PWR plant at Yankee Rowe was in line for a "lifetime extension programme" despite embrittlement.

Public opinion had it shut down.     For Good.

No secondary Containment

Neither Magnox nor AGR reactors have any form of secondary containment. The steel pressure vessel is housed inside reinforced concrete which provides a radiological barrier but which is not designed to provide any containment in the event of a pressure vessel explosion.

Corrosion of Components in the Pressure Vesel.

In 1968 it was discovered that inaccessible mild steel components in the pressure vessels of all but one Magnox reactors were being corroded by the CO2 coolant gas. In 1970 these reactors were downrated. The corrosion has been continuing at a reduced rate ever since, although it is impossible to define by exactly how much because of the inaccessibility of the components. One of the dangers of corrosion inside the pressure vessel is that parts may break off and block the flow of coolant gas, leading to overheating and potential meltdown.

Pressure vessel embrittlement

Constant irradiation and exposure to high temperatures enevitably lead to embrittlement of the pressure vessel.

Cracking and Defective Welds of Coolant Ducts.

Again, this is a fault found in several Magnoxes. Cracks up to 3 metres in length were found at Hinkley. Cracks have also been found at Dungeness.

Risk of Fire

All magnox and AGR reactors are exposed to the risk of a serious fire in the reactor if the cooling system should fail of air enter the core.

Radiation and Health

The ageing Magnox stations emit high levels of "shine", direct gamma radiation. This is because the pipes carrying the radioactive CO2 pass outside the reactor's concrete shield.

No Protection from Seismic Activity.

None of the Magnox stations were built to withstand the level of seismic shock which the NII expects of newly built stations.

A Statement from Dr. Don Arnott.

Because of the large and destructive fluxes of heat and radiation continually present in an operating reactor it is inevitable that many things in its structure suffer deterioration over time. Behaviour may alter and there may be structural weakening or actual physical damage.

In addition the graphite blocks which form the moderator in the reactor core are adversely affected.

· Changes in heat conductivity occur
· The blocks undergo a complicated pattern of expansion and contraction leading eventually to nett shrinkage in core height.
· The carbon dioxide coolant gas, activated by gamma rays, also corrodes the graphite leading to actual loss in weight. This particular phenomenon, though acknowledged as important for newer Magnox designs of higher rating (ie those with pre-stressed concrete pressure vessels, notably Oldbury) is claimed to be of lesser importance in the lower rated and older Magnoxes such as Trawsfynydd (now being decommissioned).

Nevertheless one must see the emerging pattern of mounting defects as a whole. In a way all industrial equipment resembles people: teething troubles in infancy, another set of problems in maturity and in old age an accumulation of misfortunes, no one of which may be fatal, but which can fatally interact.

Thus, briefly summarised, the pattern of graphite defect is not uniform over the whole core; structural deterioration is greatest towards the centre of the core, at greatest exposure. Such non-uniformity of damage is likely to reduce stability in the event of a mechanical shock arising for example, from duct breakage. Nor should we forget the gradual weakening induced in the graphite pile by gas turbulence: the effect of pumping coolant gas under pressure through the structure which is continuous and hardly calculated to improve structural strength.

Magnox are the oldest design of commercial nuclear power station still functioning. They have had their day and they are now obsolete: no Inspectorate in the world would nowadays license a commercial reactor that used uranium oxide fuel. This does not mean that their safety standards were necessarily unsafe from the beginning; it does mean that it is very difficult to see how safety can continue to be guaranteed in conditions of gradual deterioration with age.

British Industry has a bad habit of running equipment to death, long after it should have been shut down. It seems that BNFL are determined to do this with the Magnoxes, if allowed by the NII.

To impose it on an ageing reactor showing signs of wear and tear is potentially dangerous and there can be no excuse for incurring it.

Concern about Wigner Energy goes almost all the way back to the beginning of the nuclear industry in Britain. To sustain nuclear fission in a reactor the neutrons which emerge at a speed of about 6000 miles per second are slowed down or 'moderated' to about one mile per second. This greatly increases the probability that they will produce a fission in a further Uranium 235 nucleus.

In Britain the decision was made very early on to use graphite as the moderator. When fast neutrons collide with the carbon atoms of the graphite two things happen; firstly, they are slowed down by transferring their energy to the carbon atoms of the moderator, and secondly, part of this transferred energy is released as heat and part of it causes a displacement of the carbon atoms in the graphite crystal lattice. The graphite changes its shape and 'locks up' some of the collision energy; this is Wigner Energy.

In 1952, a spontaneous release of Wigner energy occurred whilst Windscale Pile No.1 was shut down. This alarmed the operators who then drew up a programme for the controlled release of this energy at regular periods. It was the misreading of thermocouples during one such 'controlled' release that precipitated the Windscale fire in 1957. Initially it was thought that the graphite moderated Magnox reactors would also require the periodic controlled release of Wigner energy.

It was stated in 1962 that Calder Hall and the other Magnox reactors 'would require a Wigner release in five years'. In 1967 a reactor at Chapelcross experienced a meltdown in circumstances which have never been fully explained. The accident occurred at the time of restarting. It was later claimed that Magnox reactors operated at temperatures high enough to obviate the need for such measures. There is however, a suspicion that the 'convection cooling test' that the CEGB wanted to carry out at Trawsfynydd in 1988 was linked to the need to release the Wigner energy that had built up. That test was abandoned as it bore too many similarities to the experiment that destroyed Chernobyl two years earlier.

The report into the Windscale fire revealed that there were aspects of Wigner energy which were 'not clearly understood' until after the accident. Apprehension about Wigner energy is still evident, with NIREX refusing to allow graphite from the Windscale piles to be placed in its specially designed boxes on the basis that its behaviour is still 'not clearly understood'. The fear is that backfilling the boxes with concrete, with its exothermal reaction, will by warming up the graphite precipitate a fast release of Wigner energy.

This violates the criteria set by the NII for the passive safety of intermediate level waste, one of which is that the waste should not ignite. BNFL have yet to convince NIREX, let alone the NII that they can safely condition the graphite from the Windscale reactors.

Meanwhile, BNFL insist that concern about Wigner energy, graphite degradation, and 'explosability', are not deterring them from full scale decommissioning of Magnox reactors.
No doubt we will learn more about the practical implications from Japan, where the Tokaimura Magnox reactor is to be dismantled within the next eight years.

BNFL is today announcing a lifetime strategy for its fleet of Magnox nuclear power stations. The strategy provides a phased programme for the cessation of electricity generation at the eight stations, most of which began operating in the 1950s and 1960s.

The reactors are licensed to operate for between 33 and 50 years and this early announcement of the Company's strategy for the lifetimes of the stations will allow operational plans to be optimised. For business reasons, Hinkley Point A will not be brought back into service from its current shutdown.

With today's announcement the Magnox station lifetimes will be planned as follows:-

* Continuing to run Oldbury and Wylfa to these dates depends upon the development and use of Magrox fuel. Magrox is a fuel in which uranium is used in ceramic oxide rather than metal form. A decision on the use of Magrox fuel will be taken in around 2003. Oldbury and Wylfa will also need to undergo a Periodic Safety Review in order to secure operation to these dates.

BNFL's Chief Executive Norman Askew said: "Everyone knows that these stations have a finite life and there has been speculation as to our intention regarding their operating lives.

The reason we are making this announcement today, well ahead of time, is to provide certainty about the future for all concerned. It will bring clarity to the Company's business plans, explains our plans to our employees and provides us with time to work with the communities around our stations on plans for decommissioning.

These stations were pioneers in the nuclear industry and have made, and are continuing to make, a huge carbon-free contribution to the electricity generating industry. This decision will mean that the reactors will not be run beyond the dates announced. However, both market conditions and technical issues could result in earlier closure."

The lifetime strategy announcement means that the Magnox reprocessing plant (B205) at Sellafield will close once all Magnox fuel has been reprocessed. It is expected that this will be around 2012 although this could be later depending on throughput schedules achieved. Based on the same programme, Magnox fuel production, which is carried out at the Company's fuel manufacturing site at Springfields, near Preston, will cease by 2010.

Shepway Friends of the Earth Newsletter       February 1997

Over 300% Underestimate of Radioactive Tritium
Discharged from Dungeness 'A'

Tritium is a radioactive isotope of hydrogen that readily combines with carbon, the main constituent of DNA. There is much concern that, if ingested, damage to the genes in the DNA may be the cause of stillbirth clusters, which have been found around plants that produce this isotope. Nuclear fusion, the dream, clean answer to nuclear fission produces a large amount of tritium and for this reason there is a tremendous amount of research currently being undertaken into the possible effects tritium may have on man. We await their results, but as yet we do not know!

A review of radioactive waste management at Dungeness 'A' Nuclear Power Station has revealed that discharges of Tritium are over three times higher than they were estimated to be when the present authorisation was granted in 1993.

Magnox Electric PLC, the current operators of the station have been forced to apply to the Environment Agency for a variation to their current radioactive waste disposal authorisation. The following are extracts from that application:

RADIOACTIVE SUBSTANCES ACT 1993 APPLICATION NUMBER AW9960
APPLICANT: Magnox Electric PLC, Dungeness A Power Station

"Magnox Electric PLC (the Company) has applied for a variation to the Authorisation granted to Dungeness 'A' Nuclear Power Station for the disposal of radioactive waste gases, mists and dusts. This document details the reasons for requesting this variation and the Company's proposals for that Variation.

The current Certificate of Authorisation, which came into force on 1st July 1994, limits the total quantity of Tritium that may be discharged in any period of twelve consecutive months to 2 Tbq. [Tera-bequerels of radiation]. This limit was set on the basis of past discharges while at power; it did not include any margin for air purges during outage [shut down for overhaul]. Similarly, it was anticipated that the quarterly notification level (QNL) of 320 Gbq [giga-bequerels] would be exceeded only occasionally. Recent experience at this station and others demonstrates that these values do not allow adequate room for normal operations when discharges during reactor outage periods are included.

Consequently, the QNL value was exceeded for the first quarter of 1996 and the annual limit is under significant pressure."
Magnox Electric plc Ref: WM\DNA\3.0.1
(Public consultation document Ref: AW 9960)

Appendix 2 1.2 of the consultation document states:

"The operator estimates that reactor air purges will cause maximum tritium discharges of 1.4 Tbq per year, if three reactor outages were undertaken. The tritium content in other discharges, established over many years, has been around 0.6 Tbq for the twelvemonth period and could be higher [our emphasis]. Thus the combined discharges could have a tritium content of about 2 Tbq......[300% more than previously estimated]"

NB: 2Tbq is the current legal limit above which an offence under the Radioactive Substances Act 1993 will have been committed.

This gross miscalculation should be enough to convince even the pro-nuclear lobby that the operators of Dungeness 'A' have been incompetent over many years and still do not fully understand the technology they are using!

"Do we have to wait until disaster overwhelms us before we make the changes necessary to protect our world for future generations?" JOHN GUMMER, 1994

Bill Root
Station Manager
Dungeness 'A' Nuclear Power Station
Romney Marsh
Kent
17th March 1999

Dear Mr Root

Would be kind enough to answer some questions we have regarding the Magnox Dissolution Plant. (MDP)?

(1) When does BNFL (Magnox Division) expect the MDP to come back on line?
Answer: The plant is on line now, on a regular basis.

(2) How much of the 60 tonnes of waste originally intended to go through the plant has already been through the process?
Approximately 10% of the material accumulated since the plant first operated.

(3) Has there been any revision in the amount of material expected to go through the plant, if so what is the new intended throughput figure? How much waste is still accumulated and waiting to go through the MDP?
There is no intention to process more material than originally intended. The plant always intended as a pilot project, has gone through a great deal of evaluation and development to reach the point where it is now operating regularly.

(4)
a) Does Dungeness A have a higher discharge authorisation than other Magnox plant as a result of the     siting of the MDP?
   No

b) In particular is there a higher authorisation for Cobalt-60 and Am-241?
    No

c) Does the MDP have its own separate discharge authorisation?
    No

(5) Has environmental monitoring around Dungeness found any raised levels of isotopes such as Cobalt-60 and Americium-241 which may have resulted from the MDP?
No (refer to MAFF/SEPA Radioactivity in Food and the environment report (RIFE-3, 1998)

(6) Please give the dates since 1988 that the MDP has been operating?
We introduced the plant as a pilot project and always knew that other site matters would take precedent over its operation. It has been in part time operation since 1988. The plant forms what is basically a decommissioning activity.

(7) How much of that time has the Ion Exchange System been operating?
The Magnox Dissolution Plant has not, to date, required the services of the Ion Exchange System.

(8) Has any waste been transported to Dungeness from other nuclear plants to be put through the MDP? If so how much?
None
(9) Are there any plans to build a similar plant at other Magnox sites?
We are keeping the possibility of more plants under review as one possible option for managing the waste stream. There are no firm plans at the moment.