The uranium fuel treatment facility is located, in Tokaimura (Japan), 120km northeast of Tokyo, in the Ibaraki Preface. It is currently owned by the company JCO.
The nuclear accident of the installation took place on September 30, 1999, in the nuclear plant conversion building.
The installation consists of three auxiliary uranium conversion buildings:
- One with an annual capacity of 220 tons of uranium per year for low enrichment (approximately 5%).
- Another with an annual capacity of 495 tons of uranium per year for low enrichment (less than 5%).
- Another, the one who had the accident, with an annual capacity slightly higher than 3 tons of uranium per year for high enrichment (not exceeding 20%).
In this third building, concentrated uranium oxide powder is produced from the transformation of uranium hexafluoride. It didn't work continuously. The installation was used only for very specific orders for immediate production. It was practically only operational 2 months a year.
Causes of the accident
To understand what happened first we have to briefly explain the uranium enrichment process at the Tokaimura plant.
The uranium enrichment process is performed by previously converting uranium into a compound, uranium hexafluoride, which is gaseous under normal conditions. The next step is the conversion of enriched uranium in the form of uranium hexafluoride to uranium oxide, which is achieved in a tank with an aqueous solution of uranyl nitrate.
The compound is converted by precipitation and sedimentation, and later by calcination, into ceramic fuel pellets, which will constitute the fuel elements of some nuclear reactors.
According to the established internal operating procedure, the uranium oxide (U 3 O 8 ) solution should be in a tank arranged for that purpose, then transferred to a solution of pure uranyl nitrate and homogenized with a gas nitrogen purge.
Subsequently, the mixture was poured into the water-cooled precipitation tank to evacuate the residual heat generated by the exothermic reaction that occurs.
To prevent the appearance of a criticality (a self-sustained chain fission reaction), the procedure established limits for the amount of uranium that should be transferred to the precipitation tank, a maximum amount of 2.4 Kilograms of uranium.
The work procedure was modified in November 1996, without permission from the competent regulatory authorities, allowing the treatment of the dissolution of uranium oxide in stainless steel buckets, which did not comply with the appropriate measures. This new work method had been carried out several times before the accident occurred.
Thus, when preparing the JOYO reactor fuel in September 1999, the workers dissolved the U 3 O 8 powder in nitric acid in the stainless steel buckets and poured the solution directly into the precipitation tank.
The solution used of 16 liters of uranium oxide, enriched with 18.8% of uranium-235, was distributed in four stainless steel buckets to be poured into the precipitation tank.
On the morning of September 30, when the volume reached 40 liters, equivalent to 16 Kilograms of uranium, much higher than the initially limited amount, the critical mass necessary for a nuclear fission reaction in self-chain was initiated maintained, accompanied by the emission of neutrons and gamma radiation.
Development of the nuclear accident
The worker, who added the seventh cube of uranium nitrate to the sink, saw a blue flash of Cherenkov radiation. He and another worker who was near the sink immediately experienced pain, nausea, shortness of breath and other symptoms; A few minutes later, already in the decontamination room, he vomited and lost consciousness.
There was no explosion, but the result of the nuclear reaction was an intense gamma, and the neutron radiation from the sedimentation tank, which triggered the alarm, and then the actions to locate the nuclear accident began. In particular, 161 people were evacuated from 39 residential buildings within 350 meters of the company (they were allowed to return to their homes after two days). 11 hours after the start of the nuclear accident, a gamma radiation level of 0.5 millisievert per hour was recorded at one of the sites outside the nuclear plant.
The nuclear fission chain reaction continued intermittently for approximately 20 hours, after which it stopped due to the fact that water was added to the cooling jacket surrounding the settling tank. Water played the role of a neutron reflector, and boric acid was added to the settler (boron is a good neutron absorber); 27 workers participated in this operation, who also received a certain dose of radiation. The ruptures in the nuclear chain energy reaction were caused by the fact that the liquid boiled, the amount of water became insufficient to reach criticality and the chain reaction was attenuated. After cooling and condensing the water, the reaction resumed.
Neutron radiation ceased along with the chain reaction, but for some time the dangerous level of residual gamma radiation from fission products remained in the sump. For this reason, it was necessary to install temporary protection against sandbags and other materials. Most volatile radioactivenuclear fission products remained inside the building due to the fact that they maintained a lower pressure than outside and then were collected using high efficiency air filters. However, some of the noble radioactive gases and iodine 131 entered the atmosphere.
Consequences of the accident
The accident directly affected the three operators who prepared the sample, who had to be hospitalized, two of them in critical conditions, and who died one at 12 weeks and another, after 7 months. It is estimated that one of the deceased workers was exposed to radiation of between 1 and 20 sieverts.
In addition, 56 more workers at the plant were exposed to radiation, of which at least 21 people received significant doses and had to be under medical evaluation.
Within a radius of 200 meters around the facility, access was restricted, and additionally, the Japanese authorities established evacuation measures of 161 people, from areas located at a distance of 350 meters from the plant.
As a preventive measure, the 310,000 people who lived 10 km away were advised not to leave their homes, until the situation was under control, with their confinement lasting 18 hours.
Once the criticality ended, adding boric acid to the solution of the precipitation tank, and thanks to the containment systems of the site, always in depression with respect to the outside, the levels of radiation in the exterior returned to normal.
According to the IAEA, the radiation levels of the areas near the plant, in mid-October 1999, had recovered the natural background levels. The measure of iodine-131 in soils and vegetation outside the facility, determined that the food had not been affected.
The accident was classified as level 4 according to the INES Scale (“accident without significant risk off site”), since the amounts of radiation released abroad were very small, and within the established limits, but within the site, the damages produced in equipment and biological barriers were significant, in addition to the fatal exposure of workers.
From the accident, to which all indications point as a human failure, the fuel manufacturing plants in Japan, were fully automated, to ensure that a criticality accident did not occur again, equipping the systems with neutron control equipment, and using dry conversion methods, intrinsically safer.