Wing J. Toxic vapor concentrations in the control room following a postulated accidental release

Washington: Division of Site Safety and Environmental Analysis Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission, 1979. — 44 p. An accidental release of a toxic gas can harm or even kill the individuals-who are exposed to the gas beyond its toxicity limits. Under most circumstances, the consideration of such an event would fall in the category of industrial safety. There are, however, situations that can be postulated in which a toxic gas release could interfere with the safe operation of a nuclear plant. (See General Design Criterion 19, Appendix A, Part 50, Title 10, Code of Federal Regulations). For instance, a catastrophic release of chlorine from a rail car, in which the entire load of chlorine is spilled, could lead to the incapacitation or even death of the control room operators. This is a consequence of low probability (Hornyik; Murphy et al.; Simmons et al.). A number of events may have to occur in a sequential fashion to produce dangerous levels of chlorine within the control room. The chlorine tank must rupture, the wind direction must be toward the control room fresh air inlets, and the flow conditions must be such as to allow the chlorine to reach the inlets with concentrations sufficiently high to overcome the operators. The staff at U.S. Nuclear Regulatory Commission (NRC) has concluded that, notwithstanding the low probabilities, this type of an accident scenario should be considered in its safety evaluation of nuclear reactor plants. The NRC staff has developed a number of guidelines and criteria to assure that the requirements of General Design Criterion 19 are met. These can be found in Standard Review Plans 2.2 and 6.4 and in Regulatory Guides 1.78 and 1.95. One aspect of the safety evaluation of nuclear reactor plants is the assessment of the toxic vapor concentrations that may result from a liquid spill. The present report summarizes an acceptable method for such an assessment, and lists some typical toxic substances and their pertinent physical properties. Many of the assumptions underlying the mathematical formulation in this report may seem rather simplistic and arbitrary. However, it is believed that they are reasonably conservative. As soon as more realistic models or'direct experimental data become available, the present treatment will be modified.
It should also be pointed out that the probabilistic nature of the catastrophic spill of toxic chemicals, during transportation and in storage, is not considered here. That-is, the frequency of shipment and cargo size of each toxic chemical past the nuclear plant site, the accident rates of on-site release.and of each'shipment type (trucks, rails, barges, etc.), the distribution of wind speeds and directions, and the uncertainty of the weather conditions will not be included in the assessment of vapor concentrations. The accident probabilities of trucks, rails, and barges, per vehicle mile, have been estimated by Directorate of Regulatory Standards. An analysis of the risks in the water transportation of hazardous materials
has been made by the Committee on Hazardous Materials. The statistical nature of the hazards to nuclear power plants from surface traffic accidents and from off-site release of toxic vapors has been extensively studied by Hornyik. The risk of catastrophic spill of chlorine during transportation has been investigated by Murphy et al. and Simmons et al.