Nuclear Reactor Meltdown - What it is and is not

Meltdown - What it is and is not

The term - Meltdown - refers to melting of the fuel in the reactor. Unfortunately, the term has been loosely applied to refer to any case of fuel melting, however minor. Only in several events - Three Mile Island 2 and Chernobyl - has there been significant fuel melting and only in the case of Chernobyl were there significant offsite releases.

Overheating of the fuel typically can be caused only if there is an inability to remove heat from the fuel. Two situations are the only likely causes:

Loss of coolant in the reactor cooling system followed by a failure of the emergency core cooling systems to operate
Failure of the reactor protection system to shutdown the reactor down when required for a major fault

Such conditions are considered to be outside the design basis for nuclear plants and are referred to as Class 9 accidents. The design of the plants is intended to assure that such conditions do not occur - due to the redundancy and diversity of the reactor protection, emergency core cooling, and containment isolation systems, as well as the containment structure itself. In spite of this, the US Nuclear Regulatory Commission is requiring licensees to develop procedures for such cases. The procedures are referred to as Severe Accident Guidelines.

If a meltdown does occur, a release of radioactive materials to the environment can occur ONLY IF there is also a major failure of the containment structure. For this to occur, the following would also have to happen:

Overpressure of the containment
Failure of the containment isolation systems, lines, and valves to close.

Potential causes of containment overpressure are:

Steam explosion in the reactor vessel or a dropping of at least 20% of the fuel mass of a molten core
Generation of gases either due to hydrogen generated from a chemical reaction between Zircaloy (used in the fuel cladding) and steam at temperatures above 3400F or due to carbon dioxide generated from interaction of molten core material with the concrete structures under the reactor.
Heating of the containment atmosphere due to a failure of the containment cooling and spray systems.

For there to be a meltdown with releases offsite, the following sequence would have to occur:

Failure of the reactor to shutdown when required such that it continues to produce heat at a high rate OR a major amount of coolant is lost from the reactor cooling system,
Diverse and redundant high and low pressure emergency cooling systems are unable to provide cooling to the reactor cooling system,
Fuel melting starts and blockage of flow channels occurs in the reactor such that cooling cannot be provided,
Diverse and redundant containment cooling and spray systems are unable to provide cooling to the containment atmosphere,
Redundant Hydrogen recombiners will not operate,
Containment isolation system and associated valves do not close as required,
Specialized high efficiency particulate, absolute, and charcoal filters do not function as required.

The design of the plant systems is intended to reduce the likelihood of such an event occurring (e.g. once in 250 years for the 400+ reactors with current designs). It is impossible to say, with 100% certainty, that a fuel melting event will not occur. The redundancy and diversity of plant design, NRC regulations, plant Technical Specifications, plant operating procedures and operator training and qualification provide the defense in depth.

The Chernobyl event occurred during a test and was due to a combination of design deficiencies and operator error. The event sequence included:

A power spike which resulted in a localized overpressure of the reactor cooling system, which caused a loss of coolant (This was NOT a nuclear explosion),
The power spike also causing overheating of the fuel,
Steam release with overheating of the graphite resulted in a fire, which in turn resulted in burning and dispersal of the reactor core's contents.

BWR, PWR, CANDU, and VVER designs do not have positive void reactivity coefficients as in the Chernobyl design, nor do they have graphite. These factors preclude an event of a similar type.

Click for more on less significant melting events.

References

1. Nuclear Power Reactor Safety, E.E. Lewis, Wiley-Interscience (1977), Section 9-4, page 480 et seq
2. Nuclear Reactor Engineering, Samuel Glasstone and Alexander Sesonske, Van Nostrand Reinhold Company, 3rd Edition (1981), Section 11, page 724 et seq
3. Introduction to Nuclear Power, John G. Collier and Geoffrey F. Hewitt, Hemisphere Publishing Corporation, (1987), Chapter 5, page 119-147
4. Environmental Radioactivity from Natural, Industrial, and Military Sources, Merril Eisenbud, Academic Press, 3rd Edition (1987), Chapter 14, page 343-389