The earthquake and tsunami that hit the east coast of Japan on 11 March knocked out power for a prolonged period at Fukushima Daiichi nuclear power plant. The station blackout in turn led to explosions at units 1, 3 and 4 and to serious damage of units 1-4 (units 5&6, although flooded, were relatively unscathed). As the emergency response period subsided, national and international organisations worldwide responded with reports and recommendations. On site, controlling utility TEPCO stabilised reactor cooling and, with the Japanese government, developed a long-term decommissioning plan. Below is a digest of principal events on site.
In the early days of the crisis, fire trucks pumped seawater into Fukushima Daiichi units 1-3 (as unit 4 was shut down at the time of the tsunami, its heat-generating fuel was in its spent fuel pond). Although this early cooling water was essential for reactor safety, it contributed to what is probably the biggest near-term decontamination issue on the site: the vast amount of contaminated water flooding low-lying basements of reactor and turbine buildings, and trenches (estimated at 100,000 m3 in June). Most of this was seawater that burst through ground-level doors and windows in the tsunami surge, and then became contaminated by radioactive cooling water that leaked out of the reactors’ primary containment vessels.
Starting in June, a treatment process started to feed the flooded water through a decontamination system and circulate it back into the reactors to cool them. The first Areva-Kurion decontamination system was shown to reduce concentrations of the dominant long-lived radionuclide, Caesium-137 (half life of 30 years, respectively) by one million times. In August, a Toshiba-led SARRY (simplified active water retrieve and recovery system) caesium adsorption system started up in parallel to improve the stability and reliability of the whole water treatment system; its decontamination factor reached 500,000 in late November. A second SARRY system, also in parallel, started up in October.
In the system, decontaminated water is sent to another processing unit that removes the corrosive salt in the water. A first reverse-osmosis-based unit started up in June (desalination factor 560); it was augmented by three lines of evaporative concentration apparatus in early and late August, and October (desalination factor 9000).
Initially, TEPCO planned to use existing storage tanks on site to hold all of this water, but as the extent of the contaminated water became clearer, arrays of tanks of varying shapes, sizes and contents have sprung up all around the site. High-level contaminated radioactive water tanks for 2800 tons were installed in September. Processed water tanks amounting to 145,200 tons had been installed by mid-December; units 5&6 have employed a 10,000 ton-capacity barge formerly used as a sport-fishing attraction to hold contaminated water. As of 7 February, 230,000 m3 of contaminated water had been treated, and 76,500 m3 remains inside the buildings. This is a long-term initiative; according to a recent decommissioning plan, water treatment is not scheduled to end until 2021.
Initially, the feedwater system injected cooling water into reactor units 1-3, which places water at the bottom of the RPV, and cools it through evaporation. Starting in September, TEPCO began to cool the reactor cores of units 2 and 3 from above the fuel, by injecting water through the core spray line.
The actual water supply architecture for injection is complicated, partly because of the various sources of water (offsite fresh water, onsite processed water from various tanks, onsite emergency water sources), and partly to provide redundancy and defence-in-depth in case of another accident.
During the early days of the crisis, helicopter crews attempted rather unsuccessfully to dump cooling water from a helicopter into the fifth-floor spent fuel pools. Then TEPCO switched to using concrete pumps to cool the spent fuel pools of units 1, 3 and 4 (unit 2’s spent fuel pond was also cooled with seawater injection, probably through its opened blow-out panel). By late August, TEPCO installed a two-loop alternative cooling system in the spent fuel pools of all four units. The system takes water from the spent fuel pool’s skimmer surge tank and runs it through a heat exchanger, which cools it by heating up water in a new secondary loop that dumps its heat to the air via an outdoor fin cooler. Later, desalination and decontamination circuits running through truck-mounted skids were also attached to the spent fuel ponds.
State of the core
TEPCO now estimates that unit 1 was the worst affected; most of its fuel has drained out of the reactor pressure vessel, via the bottom-mounted control rod tubes or instrumentation penetrations, into the concrete primary containment vessel. There was no water injection into unit 1 for about 14 hours. TEPCO reasons that the unit’s decay heat before seawater injection ‘significantly exceeded’ RPV water and materials’ heat absorption capacity; this fact also explains why the RPV temperature has been low from an early stage. Units 2&3 have not been as badly affected; they were each denied water for a much shorter time than unit 1, about six and a half hours. In consequence, lesser amounts of fuel from units 2&3 have dropped out of the RPV and into the PCV, based on comparisons of decay heat versus total heat absorption capacity of water in the core, and based on evidence from RPV temperature trends after water injection restarted. The right-hand image includes two cases; a best-case estimate based on data collected in March (top circle), and a worst-case estimate (bottom circle). Unit 4 had no fuel loaded in the core at the time of the incident, and so was not damaged.
There have been three reported releases of radiation into the seawater at Fukushima Daiichi. One was intentional: in April, TEPCO flushed out 10,000 tons of contaminated wastewater from the low-level central radioactive waste storage facility, to make way to treat highly-contaminated floodwaters. It also released about 1500 tons of low-level contaminated water from the units 5&6 sub-drain pits, to try to stop that water flowing into the buildings.
Radiation leaked out of a trench near Fukushima Daiichi unit 2, discovered and plugged in early April. Based on the angle of the water spouting from the leak, engineers estimated the volume of the water released as 520 m3, and estimated that the total radioactivity emitted as 4.7x1015 Bq.
To stop the leaks, TEPCO installed steel plates in the screen rooms of units 1-4, silt fences in front of the screen rooms of units 1-4, sandbags around the south pier of the power station, and 10 sandbags with radiation absorber zeolite in front of the screen rooms. TEPCO also installed floating silt screens to reduce the mobility of radioactive silt. Since then, it has also set up a pumped circulation decontamination system using a zeolite tower on the quayside.
In May, highly radioactive water was found leaking into a cable pit near the Fukushima Daiichi unit 3 water intake, found to have Cs-137 concentrations of 5x104 Bq/cm3, 1.25 million times the legal limit, but was stopped the same day.
Later, in June, TEPCO set up a larger concrete wall blocking off water intake, and installed a line of steel pipe sheet piles in the August to September period on the south end of the water intake. As of late January 2012, preparations were underway to install a permanent steel pipe sheet pile wall in front of the unit 1-4 water intakes. Environmental monitoring will continue.
In April, Japanese safety regulator NISA estimated the total amount of radiation released by the Fukushima accident in the first several weeks of the crisis was about a tenth of that released by the 1986 Chernobyl disaster, about 5x1017 Bq. That includes between 0.6-1.2x1016 Bq of Cs-137, which has a 30-year half life.
Air emissions in January were about 70 million Bq/hr, 45 million from unit 3, 20 million from unit 2 (which did not suffer an explosion) and two million from unit 1, which benefits from a temporary soft-sided frame to reduce the potential to spread contamination (see pictures below). In August-October 2011, the frame (1) was assembled by crane using remotely-actuated rigging gear to reduce the radiological risk; then panels (2) and roof trusses were lifted into place (3).
Exhaust gas (40,000 m3/hr) is filtered, sampled, and vented through the station’s smokestack. In October, the pre-filter detectors measured Cs-137 concentrations of 1.4x10-4 Bq/sec, and post-filtration values were 175 times lower. Plans are in place to install other temporary, and then permanent, covers. Work has begun to clear rubble on top of units 3 and 4.
TEPCO has used remote-controlled excavators, dump trucks, and small and large vacuum suction rigs to clear the site of the debris left by the tsunami. Site workers sprayed buildings and grounds with a green-dyed polymer spray to immobilize potentially-radioactive dust.
TEPCO’s total plan to decommission Fukushima Daiichi units 1-4 will proceed in three phases, and is expected to take 40 years. The two key tasks are removal of fuel from the spent fuel pools (beginning with unit 4, then unit 3) and fuel debris removal starting in about 2019, after intensive inspections and R&D. For now, however, the document says that TEPCO is working to a three-year plan from a safety directive produced by the government regulator Nuclear and Industrial Safety Agency called ‘Ensuring mid-term safety’. Targets and schedules will be released on an annual basis.
To date, the Fukushima Daiichi nuclear accident itself has not caused any documented deaths. However, there have been six deaths on site: two workers were killed by the tsunami, and four site workers have fallen ill and died from unrelated causes (two heart attacks, one case of acute leukaemia and one case of sepsis). The number of workers exposed to radiation to 31 December, the latest figures available, are shown (Table 1). Although an annual dose of 100 mSv was the standard radiation limit for nuclear power plant workers, the government raised the limit to 250 mSv for emergency tasks.
By Will Dalrymple, editor of Nuclear Engineering International magazine
It was very sad for me to see that the grass in the garden has grown up to my height, but at least the house itself was not destroyed by an aftershock. When I returned, the radiation of the ground in the garden was 3-5 Î¼Sv/hr. But under the gutters the radiation of the ground was 30 Î¼Sv/hr.