Did the Tokaimura Criticality Incident Create an AHR?

[Delta Force]

The Tokaimura Criticality Incident involved the improper mixture of 18.8% enriched uranium and nitric acid in a 100 liter precipitation tank that was 450 mm in diameter and 650 mm high. The tank was surrounded by a water filled cooling jacket. When enough of the mixture was in the tank a critical reaction began that continued for around 20 hours, only stopping once the cooling jacket was drained.

The materials involved, description of the precipitation tank, and the length of the incident makes it seem like the workers accidentally created an aqueous homogeneous reactor. Is that what happened?

 

[Astronuc]

The materials involved, description of the precipitation tank, and the length of the incident makes it seem like the workers accidentally created an aqueous homogeneous reactor. Is that what happened?

Assuming it was strictly a solution, i.e., without precipitate, then yes, it was basically an aqueous homogenous reactor. I don't remember the details, i.e., I don't know if they were precipitating a solid in the solution, which would make it a coupled system.

From the World Nuclear page:

They had previously used this procedure many times with much lower-enriched uranium - less than 5%, and had no understanding of the criticality implications of 18.8% enrichment. At around 10:35, when the volume of solution in the precipitation tank reached about 40 litres, containing about 16 kg U, a critical mass was reached.

That's negligent. As enrichment increases, the required volume for criticality decreases. That they applied a procedure for 5% enrichment to one with 18.8% enrichment is mind boggling. In the US, one would follow criticality guidelines, and the procedure should consider the enrichment of the U in solution. They violated basic protocols. This is why we have licenses, and only qualified personnel and institutions are allowed to have possession of special nuclear material and allowed to work with it under strict conditions.

According to the IAEA, the accident "seems to have resulted primarily from human error and serious breaches of safety principles, which together led to a criticality event". The company conceded that it violated both normal safety standards and legal requirements, and criminal charges were laid. The fact that the plant is a boutique operation outside the mainstream nuclear fuel cycle evidently reduced the level of scrutiny it attracted. The state regulator had visited the plant only twice per year, and never when it was operating.

And the accident happened in 1999.

 

[mfb]

In the US, one would follow criticality guidelines

Same in Japan.
They didn't follow the guidelines. This accident happened in Japan but people ignoring safety regulations are everywhere. Here is a list of accidents in the US, I guess most would have avoidable if all guidelines would have been followed.

 

[Astronuc]

Same in Japan.
They didn't follow the guidelines. This accident happened in Japan but people ignoring safety regulations are everywhere. Here is a list of accidents in the US, I guess most would have avoidable if all guidelines would have been followed.

I was only commenting on the Tokaimura accident.

As for the Wikipedia page, it's incomplete. With respect to the accidents in the US, the earlier accidents either didn't have guidelines, or the guidelines were incomplete. The first commercial nuclear plant, Big Rock Point, came online in 1962. Others followed.

EBR-I, SRE and SL-1 were prototype research reactors, Fermi-1 was a prototype FBR that could generate electricity. Guidelines were generally inadequate, IMO. Even as late as TMI-2 accident, March 1979, guidelines and training were inadequate, and that reactor only had about 62 effective full power days of operation in its first cycle. If it had more cycles of operation with higher burnup fuel, it would have been a lot worse in terms of fission product release. When I was discussing the Fukushima accident with a group of folks from industry days after the explosions, I don't think most knew that TMI-2 had such a short operating time and relatively low inventory of fission products until I gave them the information.

Fortunately, the US Navy under Rickover developed a rigorous program.

 

[mfb]

That was not my point. You made it sound (at least to me) as if something like this couldn't happen in the US because of better guidelines and regulations. As the history of accidents shows something like that can happen everywhere.

 

[etudiant]

Experience proves that guidelines and prohibitions eventually get violated, as was the case at Chernobyl.
Idiot proofing the nuclear industry requires making sure that no external or internal failure will cause a major release of radioactivity.
Imho, small reactors are the only plausible way to achieve that standard. The big units are just too hard to cool when things go off track.
Also, the time to build the big units is just deadly. A decade long construction cycle may be ok for the Pentagon, but not for investors.

 

[Astronuc]

You made it sound (at least to me) as if something like this couldn't happen in the US because of better guidelines and regulations. As the history of accidents shows something like that can happen everywhere.

It wasn't my intent to imply accidents couldn't happen in the US, but they are less likely since we do have regulations, guidelines and training. Some systems are better than others.

We did have two criticality accidents in 1945 and1946. Slotin violated existing protocols.
https://en.wikipedia.org/wiki/Louis_Slotin#Harry_Daghlian's_death
https://en.wikipedia.org/wiki/Louis_Slotin#Criticality_accident

Another criticality accident in 1958.
https://en.wikipedia.org/wiki/Cecil_Kelley_criticality_accident

A report on criticality accidents from 2000.
https://www.nrc.gov/docs/ML0037/ML003731912.pdf

I'm not aware of any criticality accidents in the US since ~1978. In the LANL report, in the section on the Idaho Chemical Processing Plant accident, it notes "The safety analysis prepared in 1974 identified the criticality risk if the aluminum nitrate scrub feed were to become dilute, but incorrecly assumed that stoppage of the scrub feed was also necessary. The evaluation process had been excessively focused on the physics of subcriticality and not on risk assessment."

In the Wikipedia article on nuclear power accidents by country, the accident such as the Palisades (1973) "Steam generator leak causes manual shutdown of pressurized water reactor" is an industrial accident and not necessarily unique to nuclear power plants. It is likely they didn't violated guidelines, but steam generator tubes failures were not uncommon in the 1960s and 1970s, which prompted steam generator replacements at many PWRs. Surry 2 was the first steam generator replacement ~1979/1980, and Surry 1 replaced SGs ~1982.
https://www.neimagazine.com/features/featureupdate-repair-and-replacement-trends/

On Jan. 25, 1982, UPI reported on a SG tube rupture at the R. E. Ginna nuclear plant, which is not included in the list of nuclear accidents.
https://www.upi.com/Archives/1982/0...-generator-tube-that-triggered/3917380782800/

The UPI article highlights a design deficiency, "Chronic problems with the tubes, blamed on poorly understood water chemistry and hydraulic forces, range from ruptures and cracks to thinning, denting squeezing, pitting, stress corrosion and vibration wear." Utilities were following guidelines.

The OP and my response was focused on criticality accidents, which did occur with too great a frequency, although decreasing frequency before 1978. I believe there have been some near misses though.

With regard to exposure accidents, they continue to happen, particularly outside of the nuclear power industry, but the frequency has decreased over the last 4 decades.
Reported Radiation Overexposure Accidents Worldwide, 1980-2013: A Systematic Review

 

[Delta Force]

The first commercial nuclear plant, Big Rock Point, came online in 1962. Others followed.

Even decades after the events there seems to be a lack of clarity about some of the firsts in nuclear power.

Big Rock Point was one of the first commercial nuclear reactors, but it wasn't the first in the world or even the United States (although it was the first in Michigan). Dresden predated it by several years, and it was the first commercial nuclear power plant built entirely with private financing. Bodega Bay seems to have been the first commercially viable nuclear plant proposed in the United States, but the proposed site was located close to the San Andreas Fault and the proposal was opposed by local residents and the Atomic Energy Commission.

The famous Shippingport reactor came online in 1958, but it also wasn't the first nuclear power plant to supply power to the electric grid. In fact, four reactors entered operation in 1957, including the privately owned Sodium Reactor Experiment that received Power Reactor License 1. The Sodium Reactor Experiment still wasn't the first reactor to supply power to the grid in the world (or, yet again, even for the United States).

The Calder Hall reactors began supplying power to the British National Grid in 1956, and each unit produced just as much electric power as Shippingport, 60 MWe (later 50 MWe after downrating to reduce corrosion). They were dual purpose reactors though, as illustrated by original designation as Pressurized Pile Producing Power and Plutonium, better known by the designation Magnox. The United Kingdom Atomic Energy Authority owned them and operated them primarily for plutonium production, with the power produced being a byproduct. The Calder Hall reactors were similar to the later dual purpose N Reactor operated at Hanford, the only such reactor for the United States. The plutonium producing role was so central to the Magnox design that Hinkley Point and the two following stations of the civilian Magnox reactors were built with the ability to run a military plutonium production cycle.

However, even the Magnox units at Calder Hall still weren't the first to supply electricity to the power grid. In 1955 the Boiling Reactor Experiment III reactor supplied enough power to make the small town of Arco, Idaho, the first in the world to run entirely on nuclear power, although it was only for an hour. While BORAX III was the first reactor to supply power to the grid in the United States and the achievement was widely publicized, it still wasn't the first reactor to do so in the world. Although it wasn't widely reported at the time, a reactor at Obninsk in the Soviet Union had begun doing so the year before, in 1954, and on a more consistent basis. The Obninsk reactor was still beaten by the 1951 Experimental Breeder Reactor I in becoming the first to produce electricity, although EBR-I only produced enough power to run the building it was housed in.

EBR-I, SRE and SL-1 were prototype research reactors, Fermi-1 was a prototype FBR that could generate electricity. Guidelines were generally inadequate, IMO. Even as late as TMI-2 accident, March 1979, guidelines and training were inadequate, and that reactor only had about 62 effective full power days of operation in its first cycle. If it had more cycles of operation with higher burnup fuel, it would have been a lot worse in terms of fission product release. When I was discussing the Fukushima accident with a group of folks from industry days after the explosions, I don't think most knew that TMI-2 had such a short operating time and relatively low inventory of fission products until I gave them the information.

Fortunately, the US Navy under Rickover developed a rigorous program.

Weren't emergency core cooling systems still relatively new technology in the 1970s, and something that had a poor test record as well? Are there any resources about the development of those systems?

It's interesting to note that one of the initiating events for the Three Mile Island Incident included the operators accidentally disabling the emergency core cooling system.

 

[mfb]

It wasn't my intent to imply accidents couldn't happen in the US, but they are less likely since we do have regulations, guidelines and training. Some systems are better than others.

Japan has regulations, guidelines and training, too.
Do you have a reference supporting that the US system is better than the Japanese one?

 

[Astronuc]

Japan has regulations, guidelines and training, too.
Do you have a reference supporting that the US system is better than the Japanese one?

I was reflecting on individual systems, not the collective national systems.

As for the training matters, one may consider the comment in the World Nuclear article, "It was JCO's first batch of fuel for that reactor in three years, and no proper qualification and training requirements had been established to prepare those workers for the job. They had previously used this procedure many times with much lower-enriched uranium - less than 5%, and had no understanding of the criticality implications of 18.8% enrichment."

Lack of proper training was a contributing factor during the TMI-2 accident. Post TMI-2, the NRC imposed stricter training requirements on the utilities.

From my experience, utility systems may be more rigorous than government research facilities.

 

[Astronuc]

Even decades after the events there seems to be a lack of clarity about some of the firsts in nuclear power.

Thanks for the correction; I should have checked Dresden 1 and Shippingport, which were still more experimental than dedicated commercial plant. Big Rock Point was still operating when I started working in the nuclear industry, and I had some direct experience with utility and some staff members. I had some indirect experience with Dresden 1 and Shippingport early in my career. I believe Shippingport (1957-1982) is considered the first commercial plant in the US, since although it used unique fuel and core designs, it was dedicated to electrical power production.

We can add Yankee Rowe, that operated from 1960 to 1992.
https://en.wikipedia.org/wiki/Yankee_Rowe_Nuclear_Power_Station

Commercial nuclear power plants were relatively new in the 1960s, but those designs did incorporate ECCSs. Consider the report by G. C. Lawson, Emergency Core-Cooling Systems for Light-Water-Cooled Power Reactors, ORNL, 1968.
https://www.osti.gov/servlets/purl/4825588

As for the accident at TMI-2, the comment regarding "operators accidentally disabling the emergency core cooling system," does not appear to be correct. They reduced flow from the ECCS, a mistake, since they were concerned that the pressurizer would fill completely. There were many systemic failures due to lack of training.

From the NRC backgrounder on the accident:

Unaware of the stuck-open relief valve and unable to tell if the core was covered with cooling water, the staff took a series of actions that uncovered the core. The stuck valve reduced primary system pressure so much that the reactor coolant pumps (8) started to vibrate and were turned off. The emergency cooling water being pumped into the primary system threatened to fill up the pressurizer completely—an undesirable condition—and they cut back on the flow of water. Without the reactor coolant pumps circulating water and with the primary system starved of emergency cooling water, the water level in the pressure vessel dropped and the core overheated.

From the Wikipedia article:

Once the secondary feedwater pumps stopped, three auxiliary pumps activated automatically. However, because the valves had been closed for routine maintenance, the system was unable to pump any water. The closure of these valves was a violation of a key Nuclear Regulatory Commission (NRC) rule, according to which the reactor must be shut down if all auxiliary feed pumps are closed for maintenance. This was later singled out by NRC officials as a key failure.

This part of the accident was a violation of rules/regulations. The unit should not have been operating with disabled safety-related systems. The staff clearly did not have training on how to deal with LOCA or small-break LOCA, which in this case was a stuck relief valve.
https://en.wikipedia.org/wiki/Three_Mile_Island_accident#Stuck_valve
Reference in the Wikipedia page - https://www.washingtonpost.com/wp-srv/national/longterm/tmi/stories/ch1.htm
Supplemental reference: https://tmi2kml.inl.gov/HTML/Page1.html