The Tokaimura Accident

Michael E. Ryan
Department of Chemical Engineering
University at Buffalo, State University of New York

Part II: Accident Chronology

The JCO plant needed to mix some high-purity enriched uranium oxide with nitric acid to form uranyl nitrate for shipping. The dissolving and mixing began on September 28, 1999. On the morning of September 30, 1999, three technicians, Hisashi Ouchi, Masato Shinohara, and Yutaka Yokokawa, were running fuel through the last steps of the conversion process. To speed up the process, they mixed the oxide and nitric acid in 10-liter stainless steel buckets rather than in the dissolving tank. In doing so, they followed the practice that JCO had written into its operating manual but which had not received STA approval. For convenience, they added the bucket contents directly to the 45-cm-diameter, water-jacketed precipitation tank rather than to the buffer tank. That was a crucial error because the tall, narrow geometry of the buffer tank was designed to preclude the onset of criticality. In filling the precipitation tank, the crew added seven buckets, amounting to a total of about 16 kg (35 lbs of enriched uranium, or roughly seven times more uranium than permitted under the STA license.

The three technicians were working in a small processing bay. Masato Shinohara stood on a platform and was pouring the uranyl nitrate solution into the precipitation tank while Hisashi Ouchi held a glass funnel in an inlet at the top of the tank. The third technician, Yutaka Yokokawa, was seated at a desk approximately 4 meters (13 feet) away from the precipitation tank. At approximately 10:35 a.m. the technicians added the seventh bucket and saw a blue flash. The two technicians near the vessel began to experience pain, waves of nausea, some difficulty in breathing, and problems with mobility and coherence. The gamma radiation alarms activated immediately. The blue flash that they had seen was a result of the Cherenkov radiation that is emitted when nuclear fission takes place and ionizes air. The addition of the seventh bucket had caused a self-sustaining chain reaction. The mixture, in other words, had gone critical. Mixing in the precipitation tank caused the fissile uranium species to disperse so that the reaction fizzled out. However, the critical mass later reassembled, initiating another chain reaction that released more neutrons and gamma radiation. This cycle was repeated several times over many hours.

None of the three technicians realized what had happened. Mr. Ouchi had been draped over the top of the tank and was experiencing the greatest difficulty. The other two workers helped him out of the building in response to the gamma radiation alarms. A worker in an adjoining building noticed the injured and confused technicians and called for medical assistance. An ambulance arrived quickly and removed the affected workers. No one at JCO initially understood the nature of the problem and management eventually requested help from other nearby technical facilities. Uncertainty existed among workers and management about evacuation of the premises or other action that should be taken. Workers were given the option to remain at their jobs or to go home.

The exact critical mass for the 18.8% uranium mixture in the JCO precipitation tank is not known. In the Joyo reactor, the minimum critical mass for the solid 18.8% uranium fuel is about 46 kg (101 lbs). But the critical mass is greatly reduced when the fuel is in solution because water acts as a "moderator." Light atoms such as hydrogen slow the neutrons released by decaying 235U nuclei between fissions, making it more likely that they will be absorbed and trigger another nucleus to decay. The critical mass was further reduced at Tokaimura because a water jacket surrounding the precipitation tank reflected neutrons back into the tank.

At 2:30 a.m. on October 1, 1999, JCO staff recognized the need to bring the precipitation tank under control and initiated a plan to drain the cooling jacket. Draining water from the cooling jacket would cause the reaction to cease. This proved difficult and required workers to dismantle pipes leading from the jacket. The workers could work in the irradiated building for only a few minutes at a time. The jacket was later purged with argon. Boric acid (effective for absorbing neutrons) was pumped into the tank in order to mitigate the chain reaction. The reaction was arrested at approximately 6:00 a.m. on October 1. Judging from the levels of gamma and neutron radiation measured near the plant perimeter, the criticality excursion seems to have lasted about 20 hours. After that time, the radiation levels dropped below detection limits.


  1. What mistakes did the technicians make that resulted in this accident?

  2. Were there any flaws in the equipment or design of the process that contributed to the occurrence of this accident?

  3. What steps could JCO management have taken to eliminate or reduce the possibility of this type of accident?

Go to Part III