JCO Criticality Accident:
How Many Lessons Will It Take?
Rushed Interim Report Plays Down the Accident
The Interim Report
On November 4th, the Science and Technology Agency (STA) released data on the estimated scale of the September 30th accident at the JCO Tokai plant. The STA also announced the estimated radiation dose received by the local residents, based on this data. The following day an interim report with urgent suggestions reflecting those results was presented at the fifth meeting of the Uranium Conversion Facility Plant Criticality Accident Investigation Committee of the Nuclear Safety Commission (NSC).
The STA has asserted that about 1mg of uranium fissioned in the criticality accident. This amount is extremely close to the amount provisionally calculated by CNIC on the basis of data gathered before November 3rd by the Japan Atomic Energy Research Institute from analysis of the uranium sample collected from the precipitation basin by JCO employees on October 20th. However, when the sample was collected, the equipment set up to stir the contents of the precipitation tank failed to operate, and the sample was taken without the contents of the solution being stirred uniformly. Thus the sample represents only the top layer of the solution. CNIC is critical of the way this sample was collected, and there is a possibility that a higher amount of uranium underwent fission.
The STA also released results of tentative "theoretical" estimation on the exposure level of the area around the site of the accident. These results revealed that by the time criticality stopped at 6:15 a.m on October 1st, at a point 80 meters away from the JCO plant, where the closest public street lies, radiation would have been 160 times higher than the annual exposure dose limit of 1mSv for the general public. The radiation would have been 13 times higher than the annual exposure limit 200 meters away from the site, and over two times higher 350 meters away where evacuation was advised. It must be pointed out that this comparison is based on the annual dose limit, and residents and employees were exposed to amounts between two and 160 times the annual limit in a matter of hours. We assume that a considerable number of residents were exposed to radiation levels close to these figures by remaining too long around the site, or failing to evacuate at an earlier stage.
STA's Claims on Possibility of Cancer
The STA asserts that exposure under 200mSv will not result in cancer; however, nothing can be further from the truth. This assertion was based on exposure data from Hiroshima and Nagasaki A-bomb victims, yet in 1999 the International Commission on Radiological Protection made it clear that this data was not fit for evaluating exposure to a relatively low radiation dose with long-term effects. When exposed to radiation, varying health effects are seen according to the exposure dose. The possibility of contracting cancer increases in accordance with the exposure dose. However, even with a relatively low dose, an exposure to radiation will increase the possibility of contracting cancer no matter how small the amount is. There is a possibility that there will be adverse health effects for the local residents in the future, and these people must have a guarantee of sufficient aftercare over an extensive period.
The Accident Investigation Committee appealed for the revision of safety inspections and the establishment of safety at nuclear-related work sites. Upon receiving this request, the NSC began on November 11th the revision of the Safety Review Standards regarding nuclear fuel facilities. There is a need for an independent critical analysis of this accident. Thus CNIC is preparing to set up a Committee for Comprehensive Social Impact Assessment of the JCO Criticality Accident. Various aspects of the accident will be evaluated and a final report will be put out.
Effects Due to Neutron Release
On October 20th, the Japan Atomic Energy Research Institute (JAERI) staff collected a sample from the uranium solution left in the precipitation tank after the accident (see previous article). According to the results from the analysis of the sample, the total number of fissions was estimated as 2.5x10E+18, which corresponds to the fission of about 1mg of uranium 235. The number of fissions from 1g of uranium was estimated as 1.44~1.55x10E+14, and the amount of uranium solution put into the precipitation tank was estimated as 16.6kg.
The Science and Technology Agency (STA) mapped a graph based on the above analytical data and the measurement data of neutrons and gamma rays from the area monitoring posts (Figure 1). This is a total effective dose equivalent for the exposure of a person at a certain time and area who remained there without shielding. The results show that residents living within a 350m radius of the JCO plant were exposed to neutron dose of over 1mSv immediately after the accident. The radiation dose limit for the general public is 1mSv per year. Thus in just a matter of hours following the accident, the residents were exposed to an amount exceeding the dose limit for a year. This result has also confirmed our assertion that the evacuation area of 350m radius was set too small. In addition, this result is only an estimation based on uncertain assumprions, and the actual rate of exposure could well be higher.
There is no reliable data available of the radiation dose at the site of the accident. On the basis of past incidents, however, it is thought that the radiation level rises rapidly when one approaches the spot. The employees who extracted the coolant water 16 hours after the accident were thus forced to work under radiation level as high as about 2Sv per hour.
Fission reaction consists of an intial burst and a prolonged plateau due to delayed criticality. Based on the estimation of the National Institute of Radiological Sciences (NIRS) that the three employees at the site of the accident were each exposed to a radiation dose of about 17, 10, and 3Sv, it can be assumed that an extreme amount of neutrons were released at the time of the initial burst. On the other hand, a neutron dose rate of 6.3 µSv/h was measured immediately after the accident at a monitoring post of JAERI located about 1.7km from the JCO plant. In addition, sodium 24 was found at an elementary school 3km away from the plant. It has been calculated from the analysis of the uranium solution sample that the total effective dose equivalent received by a person 1.5km away from the site was 0.22µSv on September 30th at 11:00, and 0.45µSv on October 1st at 6:15. A significant level of gamma ray was released as well. Around 11:00 on the day of the initial burst, 0.84mSv of gamma ray was measured at a monitoring post 110m away from the site of the accident. The measured gamma ray most likely resulted directly from the criticality and was released into the environment. Furthermore, even an hour after the accident, the dose of gamma ray rose for about 20 minutes at a facility 7km west of the JCO plant. Such rises of gamma radiation were measured at a total of 38 monitoring posts. It can be assumed that gamma rays measured further away from the JCO plant were released from fission products released into the environment during criticality.
The release of radioactive materials
Many radioactive materials were produced from the neutrons released into the environment. Sodium 24, which was often talked about during this accident, is produced when a neutron hits sodium 23, itself a stable material naturally existing in the environment. Sodium 24 has a half-life of 15 hours and releases beta-ray and two gamma rays as it disintegrates. As a result it becomes stable again fairly quickly. Sodium 24 was also found in the vomit of one or two of the three employees directly connected to the accident. Furthermore, it was detected in nearby soil samples and in salt left in a nearby house. Activated gold, copper, and manganese were found as well. In addition, massive amounts of rare gases such as krypton and xenon were produced due to fission. It was learned later that nuclides such as strontium 91 and cesium 138 were detected, respectively, as decay products of krypton 91 with a half-life of 9.5 hours, and xenon 138 with a half-life of 32 minutes. It can be assumed that a massive cloud of highly concentrated radioactive rare gas was released at the time of the accident since cesium 138 was measured even a couple hours after the accident despite its short halflife, and an amount close to the concentration limit within the atmosphere was measured. In addition, radioactive iodine (iodine 131 and 133, with half-lives of 8 days and 20.8 hours respectively) were found from soil samples from various places several days after the accident. Iodine 131 will accumulate in the thyroid gland and cause internal exposure when inhaled or when consumed by eating contaminated vegetables. It was belatedly found out on October 11th that such iodine was being released into the atmosphere through a ventilating system that had been left on after the accident until that day. Radioactive nuclides were found from 108 of the mere 138 cases of soil sampling conducted by the government. In addition, nuclides were found on vegetables and wild vegetation as well. Only 17 samples of drinking water and 31 of well water were examined, yet the Nuclear Safety Commission (NSC) put out a 'safety declaration' at a very early stage of the accident.
Widespread contamination by nuclear materials was prevented since the velocity of the wind had been low and it had rained shortly after the accident, conversely exposing local residents to higher levels of radiation. However, the spread of radioactive materials cannot be predicted only on the basis of limited information derived from meteorological conditions. For instance, gamma-ray thought to be emitted from released fission products was measured even at a point 7km west of the JCO plant.
The total number of fissions estimated from the uranium solution sample was 2.5x10E+18, as mentioned earlier. This number nearly corresponds with the calculation done shortly after the accident by CNIC. Taking all these facts into consideration, above all when we think of the seriousness of the neutron exposure level outside the plant boundary, CNIC firmly believes that the accident belongs to level 5 of the International Nuclear Event Scale.
Because the JCO plant did not have any counter-measures in the event of criticality, this accident brought about radiation exposure from a continuous release of neutrons into the atmospher, and from a large amount of fission products which escaped out of the conversion plant even after the evacuation order was called off. The government's poor safety control and accident management brought about serious damage, and the cost was paid by the local residents. (By Tadahiro Katsuta)
The Science and Technology Agency (STA) originally stated that 49 people had been exposed in the accident at Tokaimura, but it was announced at the second Uranium Conversion Plant Accident Investigation Committee of the Nuclear Safety Commission that the number was now 69--i.e. 59 JCO workers who were at the plant at the time of the accident, (including the three seriously exposed), three ambulance workers who attended at the scene of the accident, and seven people who happened to be nearby. The Committee said that it based these figures on inspections of the film badges worn by workers at the site. However, the actual number of people exposed is unlikely to be as low as the figures suggest.
The publicised figures do not include 18 workers who carried out the extraction of coolant water from around the precipitation tank to stop the criticality condition, or the workers who were engaged in mounting aluminum-filled bags around the conversion test plant to shield radiation. Since these workers knew that they would be exposed, they are considered as 'planned exposure workers', and are therefore not included in the count--an odd form of calculation, to say the least. Surely all information relating to exposure in this accident should be made public as soon as possible.
The three workers who were seriously exposed are estimated to have been subjected to between 3 and 17 Sv of radiation, and two of them were actually exposed to a lethal dose. Quite a few exposed workers, including those three, were not even wearing the required film badges (integrating dosimeters for gamma ray measurement) for workers dealing with radiation. This is just one instance of the kind of irresponsibility revealed in this case.
Only a limited number of people were examined for exposure to gamma radiation or internal contamination through inhaling radioactive substances. Thus the exposure doses were measured by the following methods; calculate exposure dose from changes in the number of lymphocytes, calculate the amount of sodium 24 in the blood, check for change in chromosome, speculate on the exposure dose from the sodium 24 count measured by the whole body count (WBC), and so on. If you are exposed to neutron radiation, sodium 23 in the body was irradiated and becomes radioactive sodium 24. Then gamma radiation counts from sodium 24 can be counted by a WBC. However, the half-life of sodium 24 is only 15 hours, so the measurement has to be done right away before sodium 24 ceases to exist.
At the time of the accident there were 124 workers within the JCO property. Even the workers without film badges could have been checked for sodium 24 if they had been given a WBC sooner. The same may be said for residents living in the area of the plant. The only way to estimate the amount of radiation to which these people were exposed is to check the dosimeters worn by the 'planned exposure' workers. The ratio of neutron and gamma radiation contributions to dose equivalent was regarded as about 10 to 1. The National Institute of Radiological Science, which examined the three seriously exposed workers, estimates that 85% of the exposure was due to neutron and 15% due to gamma radiation.
According to the measurements (absorbed dose) of 44 people, including JCO workers on site by WBC, out of 18 workers who carried out the extraction of coolant water, the highest doses were found in two workers who took pictures of the coolant valve. During this three-minute task, one of them was exposed to about 120mSv (neutron and gamma) and the other to 98mSv. This means that the first person had more than 100mSv, the maximum permissible dose at the time of emergency. Another five workers had 74 to 56mSv and four others had more than 40mSv.
The estimated exposure dose of the people other than those 'planned exposure' workers is only given in the absorbed dose (Gy), a measurement unit that shows the amount of energy absorbed in exposure to neutron and gamma radiation. Proper calculations should be made so that the effects on the human body are clear. Neutron and gamma radiation counts should be made separately and be given by the unit of Sv. Exposure dose becomes more and more ambiguous as time goes by, and thus accurate data on exposure doses should be released as soon as possible.