Start of the end of the German nuclear line3 August 2002
The German government has decided to phase out nuclear power as quickly as possible without financial compensation. Based on this, the Internationale Länderkommission Kerntechnik (ILK) has studied this assessment from a scientific and engineering point of view.
The German government has deemed it necessary to phase out nuclear power as soon as possible without financial compensation. It has justified this stance based on an unfavourable assessment of the risk-benefit position.
The Federal states of Baden-Württemberg, Bavaria, and Hesse commissioned the Internationale Länderkommission Kerntechnik (ILK) to scrutinise this assessment, as well as the underlying arguments, and to include the international state of knowledge and scientific debate in so doing.
ILK concentrated on assessing the safety of German nuclear power based purely on engineering and scientific considerations. It did not take account of political considerations, nor did it consider alternatives.
The safety or risk assessment rests on the quality of and adherence to conservative design requirements, as well as the inclusion of results from probabilistic risk analyses. Consideration has also been given to empirical findings. ILK also comments on the significance of the ageing of German nuclear plants in terms of safety and on further developments that have occurred in safety engineering.
ILK compared the risk of lethal accidents by operating nuclear power plants with the risk from operating other energy producing systems.
Basic safety philosophy
The design criteria for the current generation of light water reactors were developed in the 1960s and 1970s. The protection of individuals, society and the environment were of paramount importance. These concerns led to the conservative principles of a defence-in-depth and safety margins that formed the cornerstone of reactor safety philosophy.
Even though nuclear plants are very conservatively designed against accidents, measures are still taken against events whose occurrence is deemed to be extremely unlikely or where a failure of the safety system is assumed. These additional features are intended to prevent or substantially mitigate the potentially grave consequences of a major accident and to reduce the residual risk as far as possible. Initially, only single, selective measures were taken. The evolution of the safety philosophy led to an extension to an additional safety level compared to the original design.
Uncertainties and probabilities for the failure of components cannot be quantified for this deterministic approach. Therefore systems, structures and components (SSCs) are designed in such a way that the stresses remain far below the limits at which damage can occur. For design-basis accidents, a mastery of incidents is demanded even under the stipulated condition that the staff does not intervene for the first 30 minutes after an accident.
This conservative safety philosophy has essentially remained unchanged throughout the use of nuclear power. The technical means for its implementation, however, have advanced significantly.
Improvements over time
Safety has been continuously improving over time.
• The demands placed on facilities constructed at a later date have been continuously increased, and include a degree of redundancy and the avoidance of dependent failures. Plants that were already operating were comprehensively readjusted to the new standard.
• A further focus of the increased safety reserves concerns extreme circumstances. On the one hand, they include precautions against extremely rare incidents such as a plane crash. On the other hand, on-site accident management measures have been introduced, particularly over the last decade. They enable the prevention of core melts even in the case of extensive outages of safety systems. Special emphasis was placed on taking measures for avoiding failures of the containment that might lead to large releases of radioactive substances.
• Improvements based on operating experience. These range from detailed improvements such as optimised tests and maintenance measures to changed system operating modes, use of more reliable components, up to large and small backfits, such as the exchange of piping in BWRs.
• Operating manuals used by staff during normal operations and during operational malfunction were continually improved.
• The training of staff evolved and improved. Not only content, but also training methods has been improved. In particular, the training given to shift staff who require authorised licensing to carry out their duties. Simulator training plays an important role since it provides experience in how to proceed during accidents that staff would be unable to gain during normal plant operations. Simulator training covers normal and malfunctioning operating conditions, as well as beyond-design basis conditions that necessitate use of preventative accident management measures.
The safety of German nuclear power plants is regularly assessed. On the one hand, proof must be given that the essential system functions necessary for the safety of the plant are working correctly. On the other hand, it has to be shown that the quality features derived from the quality requirements are maintained over time.
Technological ageing refers to changes in the characteristics of technical equipment over the course of operational use until their decommissioning. These changes in characteristics are generally not positive, and are taken account of in the design and operation of a plant.
To verify that sufficient margins are always available and to thereby maintain safety, ageing effects are monitored. Examples include:
• Putting radiation sample inside the reactor pressure vessel to determine in advance the impact of neutron radiation on the material characteristics and to respond with corresponding measures.
• Periodic tests of SSCs that are important from a safety engineering point of view.
• Continuous determination and assessment of the mechanisms influencing ageing, such as temperature, transients, water chemistry, and vibrations.
The results of these supervisory measures provide the basis for preventative maintenance, repairs or the replacement of SSCs.
Approximately DM 3 billion per year is spent on maintenance, periodic tests, replacement of components and backfits for all currently operating German nuclear power plants. This ensures that even those ageing effects that cannot be anticipated for the entire operating life do not lead to a decrease in safety.
Characteristics that are not amenable to periodic tests are checked by qualification of these SSCs. This includes tests of artificially aged SSCs under accident conditions. The reliability of the qualification over long time periods is verified by testing representative SSCs.
The technical run times of nuclear power plants is determined by the safety and reliability required of SSCs. The actual run time of a nuclear power plant depends on its profitability, which is decisively influenced by the costs required to maintain the safety standards and to implement upgrading measures.
The US NRC sponsored a major study that collected data and assessed the impact of ageing. The results of this programme were a major input to the formulation of the License Renewal Rule in the USA. Nuclear plants in the USA are licensed for 40 years, and the licence can be renewed for an additional 20 years.
According to the IAEA, worldwide cumulative operating experience of 9384 reactor years had been achieved by the end of 1999. In Germany, 590 reactor years of experience with light water reactors had been gained by the end of 1999. As a result, it can be said that:
• The conservative safety philosophy has been successful in that there have been no accidents in German light water reactors that have resulted in damage to the public health or the environment. All incidents have been controlled by the original design. Thus backfits are pre-emptive increases of the safety margins.
• Over the last two decades, there has been a clear decrease in the number of occurrences of important transients, such as the loss of the main heat sink, and malfunctions of the feedwater and auxiliary power supply The number of unscheduled reactor scrams has fallen over this period by a factor of 3. The decline in the number of reportable events as well as the very high availabilities of the plants also point in the same direction. This development can largely be attributed to the steady rise in quality both in terms of technology and staff training.
• The unavailability of single trains of safety systems during periodic tests has been at a very low level since the 1980s.
German operating experience is comparable with other countries with a similar safety philosophy (Belgium, Finland, France, UK, Japan, Netherlands, Sweden, Switzerland, Spain and the USA). In over 350 light water reactors of Western design and operating practice, there has been one accident involving core damage at Three Mile Island, which did not result in a serious release of radioactivity beyond the plant.
The nuclear industry and regulatory authorities initially handled unquantifiable uncertainties in deterministic reactor safety analyses by implementing defence in depth and safety margins.
In the 1970s, risk assessment was used to quantify the uncertainties, thus leading to a more rational approach to safety management. This methodology has changed the approach to reactor safety in two ways:
• The plant is analysed as an integrated system consisting of hardware and plant personnel.
• Quantitative values characterising the risk are defined and calculated. The most commonly used metrics are the core damage frequency (CDF); the large, early release frequency (LERF); the probability of death of an individual living near the plant; and the probability of a number of deaths in society at large.
Periodic safety reviews are carried out every 10 years in Germany. These are probabilistic analyses that focus on accident sequences that could lead to core damage, their basic causes, and their frequencies. It also includes examining the active safety functions of the containment.
Assessment of risks of radiation exposure
The evaluation of the health risk of exposure to ionising radiation is an important element in the risk assessment of nuclear energy. Between 1972 and 1988, a risk co-efficient of 1.25% per Sv was unanimously used for calculations. The coefficient signifies that an additional dose DD results in an additional cancer risk DR, regardless of a person's prior radiological exposure:
DR% = 1.25 x DD per Sv
This linear approach represents a practical simplification. The risk was considered to be real for DD > 200mSv, and to be hypothetical for smaller doses.
The re-evaluation of the data from the Hiroshima and Nagasaki atomic bombs resulted in an unexpectedly steep rise in the number of radiation-induced cancer cases. As a result of this, and a change of the extrapolation model, the risk coefficient was modified to 5% per Sv, and the validity of this relationship held to a value as low as DD = 100mSv. This has subsequently been confirmed.
An effective dose of 50mSv per year for workers is laid down as the limiting value for the whole body dose. This value will be lowered to 20mSv per year as a result of reassessment.
Continuous improvements in radiation protection have contributed to the continual decline in doses to staff at German nuclear plants. Achieving the new value should not present a problem to the plant operators.
For PWRs, the decrease has largely been achieved by using steel alloys with a low cobalt content. For BWRs, the reduction is due to the modification of internal circulation pumps and the replacement of piping with a reduced test effort for weld seams. In contrast, this extensive replacement was the reason for the dose peaks in the early 1980s.
In all nuclear plants, increased use was made of local shields (lead mats). In addition, the ALARA principle was implemented with greater consistency.
Benefits of nuclear energy
Nuclear energy should not be viewed in isolation, but must be compared with realistic alternatives. Furthermore, risk is only one evaluation criterion. If compatability with the precept of sustainability is being investigated, other criteria also gain relevance. Of these, the emission of greenhouse gases is the most important.
Nuclear energy has a 17% share in worldwide electricity production, but only covers 6% of the primary energy demand. In Germany, 19 reactors provide around 33% of domestic electricity production.
The share of fossil energy carriers in the worldwide primary energy turnover is well above 80%. This will also be the case in 2020, which will increase CO2 emissions from 21.3 billion tonnes to 39.5 billion tonnes. In Germany, a decline in the use of nuclear power will largely be compensated for by an increase in consumption of natural gas. This would make the German CO2 reduction target (-25% compared to 1990) unachievable.
According to the current state of knowledge, the different energy carriers and chains can be assessed in terms of their benefits and disadvantages. The benefits of nuclear energy lie in its largely emission-free electricity production, the safe availability of resources and - with reservations - in its competitiveness. The large damage consequences of 'worst possible reactors accidents' without consideration of their extremely low probability have contributed to negative public perceptions. Finally, the economic competitiveness of current generation nuclear plants could be significantly improved by a more rational management of their resources.
In the meantime, the state of knowledge concerning risks has progressed to the extent that it is possible to carry out a comparative classification of the risks of different energy supplying options.
In order to account for the special nature of nuclear risk, acute and also latent fatalities due to late radiation-induced cancer must be included in the model. This leads to a corresponding accident risk of 10-3 - 10-1 fatalities per reactor year. In comparison, there are over 210,000 cancer deaths per year in Germany.
In the assessment of accident risks, long-lasting and considerable loss of land due to contamination must be considered in the case of nuclear energy. Its extent and duration is determined by the decay time of the key nuclides, as well as by the ability to finance decontamination measures.
These problems cannot be solved by giving up nuclear power plants domestically and continuing operation of nuclear plants in neighbouring countries. Imports of CO2-emission-producing electricity are a step in the wrong direction due to their contribution to the climate changes that are an increasing cause for alarm given the current state of knowledge on their unlimited reach in terms of geography and time. A global conceptual approach also includes considering that the German safety culture and technology has had a positive influence, especially in those countries willing to construct and operate nuclear power plants which thus far still have a reputation of low existing safety standards. This positive influence would be lost should the technology be abandoned domestically.