Date: 3 January 1961
Location: SL-1 reactor, National Reactor Testing Station, Idaho, USA
Type of event: criticality excursion in research reactor
Description:
The SL-1 reactor was a prototype of a reactor intended for easy assembly at remote facilities such as DEW line stations in the Arctic. It used 15 kg of uranium fuel (enriched to 91% U-235), was water moderated, and had a thermal power capacity of 3 MWt. Five aluminum-clad cadmium control rods provided reactor control. The SL-1 had operated 2 years, with an 11-day shutdown for maintenance being completed at the time of the incident.
Three workers were reassembling the control rod drives on 3 January in preparation for startup the following day. At about 9:01 PM the three workers were on top of the reactor when one manually removed the center control rod as rapidly as possible, over a 0.5-second period. The reactor became supercritical, with a total energy release of 1.3 x 108 joules (comparable to 30 kg of TNT), producing a steam explosion. The worker who extracted the rod was killed instantly, impaled on the building's ceiling by a control rod. The other two men were burned and thrown by the steam explosion, one dying instantly from impact with a shielding block and the other sustaining head injuries of which he died 2 hours later (maximum dose sustained was possibly 350 rad). The release of radioactive material was largely contained to the building.
Emergency responders were alerted by an automated alarm and arrived at the site at 9:10 PM. High radiation readings were measured in the reactor building, delaying entry. At 10:50 PM several responders and contractor personnel removed one man alive, who died shortly afterwards. One body was removed from the reactor building on 4 January and the other on 9 January. Of personnel/responders involved, 22 received doses of 3-27 rads from entering the building and/or handling the casualties.
The reason that the control rod was withdrawn is unknown, since none of the workers survived and the facility did not have appropriate data recording systems. The control rods in SL-1 had some tendency to stick, sometimes causing difficulty during manual extraction. One hypothesis is that the worker accidentally withdrew the control rod too far in an effort to overcome a stuck condition. The amount of withdrawal involved was about 50 cm, possibly difficult to achieve accidentally, and the particular control rod involved had not been sticking for the past six months. Another hypothesis is that the rod was intentionally withdrawn in an act of murder-suicide; this was the conclusion of the investigation of the incident.
Consequences: 3 fatalities, all from mechanical/thermal effects of the explosion.
References:
© 2004-2006, 2007 by Wm. Robert Johnston.
Last modified 17 October 2007.
Return to Home. Return to Nuclear Weapons Resources. Return to Database of radiological incidents and related events.