As with war, the first casualty of a nuclear emergency is truth. The Fukushima Daiichi nuclear plant is said to have weathered the 9.0 magnitude earthquake pretty well, but things began to go wrong as the tsunami swept through. E&T sifts the evidence to find out what went wrong – and what went right.
Just as nations worldwide had accepted nuclear power was key to tackling climate change, nuclear disaster struck. What went wrong at the Fukushima Daiichi plant?
The crisis was principally a result of flooding. When the earthquake hit, the three nuclear reactors operating at the plant did exactly as they were designed to do: they shut down. With off-site power wiped out, emergency diesel-generators kicked in and started pumping cooling water to the fuel rods in each reactor. These cooling systems, designed to keep reactor containment vessel temperatures to around 260°C, operated exactly as they should – until one hour later when the tsunami swept through.
The 14m high, crushing wall of water breached coastal sea walls, flooded the power plant and its transformer yard shorting out all access to the off-site electrical grid. It overwhelmed the diesel generators at the base of the plant and irretrievably damaged many of the pumps used to transfer cooling water to the reactors. Local flooding and earthquake damage prevented outside assistance and the Japanese government declared a nuclear emergency.
For plant workers, this was 'an unimaginable situation', says Malcolm Grimston, associate fellow of energy and environment at UK-based independent international affairs organisation, Chatham House. 'Here was a group of people, worried about radiation on the site and wondering what's happening to their families,' he stresses. 'At this point it was very difficult to argue that there was any control.'
The day after the earthquake and tsunami, temperatures in reactor vessels topped 1,200°C and the explosions started. Searing temperatures saw zirconium alloy in the fuel rod cladding reacting with steam, filling the vessels with hydrogen.
Professor Tim Abram, professor of nuclear fuel technology at UK-based Dalton Nuclear Institute, says this was hardly a surprise. 'In any serious nuclear accident, one clear disadvantage of zirconium alloy cladding is it corrodes, as any metallic cladding would,' he explains. '[Fuel rod and cladding] temperature goes up, water gets reduced and the alloy releases hydrogen. It happened at Three Mile Island, Chernobyl and, as far as we can tell, here. A lot of damage is done by hydrogen explosions.'
Despite attempts by plant workers to vent this hydrogen and reduce pressure in the containment vessels, a first explosion in reactor 1 destroyed the upper cladding of the outer housing or secondary containment. Although the reactor's primary containment vessel remained intact, the Japanese government evacuated some 200,000 people living within 20km of the plant amid growing evidence of fuel rod damage and fears of radiation (see 'What happens when cooling systems completely fail', below). At the same time, residents within the next 10km were advised to stay indoors.
Two days after the first explosion, on 14 March, a second took place at reactor 3, the only reactor on the site containing mixed oxide fuel, a mixture of plutonium and uranium reprocessed from spent uranium and more toxic than standard uranium fuel. Again, the reactor's secondary containment was damaged but the explosion was larger than the previous and felt 40km away.
Finally, on 15 March, a third explosion took place at Reactor 2 that damaged its suppression chamber. The chamber, a torus structure containing a large body of water, sits beneath the reactor and is crucial to keeping reactor vessel pressure within design limits. If a reactor's primary cooling system fails, this water can get pumped into the reactor, through cooling circuits, and recirculated back to the torus. During this cooling process the water can also act as a 'scrubber', and remove the more volatile radioactive material from the fuel rods, which will be returned to the suppression chamber. Damage the chamber, and the water with its radioactive material can be released.
Grimston believes a radioactivity spike around the time of the explosion could indicate the pool was damaged and contained radioactive material. 'I think that when the pool was breached, the backlog of radioactive material that would have been scrubbed with the water could have been released into the environment,' he says. 'We saw a big spike in radioactivity which certainly went away rapidly, consistent with a one-off release.'
Above the water-line
Following the spike, radiation levels at the plant exceeded the legal safe limit and 750 non-essential workers were evacuated leaving a 'skeleton crew' of 50 employees. Japanese Prime Minister, Naoto Kan, told the men, most likely older workers beyond reproductive age: 'You are the only ones who can resolve a crisis. Retreat is unthinkable.'
In hindsight, a key question is why didn't the nuclear plant designers or utility, the Tokyo Electric Power Company, TEPCO, place the back-up diesel generators above the water-line? Wouldn't this seemingly simple decision have prevented the engines from flooding and kept them running, potentially averting the nuclear disaster?
Grimston agrees: 'A lesson could be as trivial as don't put your diesel generator at ground level. But the real question is what other trivial and obvious things are there; who's not thinking outside the box here?'
Abram asserts: 'Yes, there is a case to be made for looking at the location of these generators, but, even if they hadn't been flooded, many of the pumps the generators were supposed to power may have been flooded and ground-shorted.'
In such an event, the cooling system would have still failed, yielding the same end-result; dangerously hot fuel rods. As Abrams points out, the diesel generators were an obvious weak link, but not the only weak link. 'The fundamental fact of the matter is the plant was flooded by a tsunami well above what the plant was designed to cope with,' he adds.
Looking at the plant's specifications, Fukushima Daiichi's plant designers, TEPCO, and the Japanese regulator, did not expect 11 March to take place. The coastal plant was designed to withstand tsunamis up to 5.7m high and earthquakes with moment magnitudes of up to 7.0. The reactors survived the 9.0 magnitude earthquake but were overwhelmed by the 14m tsunami. So why wasn't the plant designed to higher specifications?
The answer is cash.
'What is a reasonable investment to make regarding the safety of that unit compared to society's other needs, such as road safety? Which would be the better payback in terms of lives saved?' asks Abram.
Some 40 years ago, Japanese regulators determined the size of earthquake and tsunami the plant should withstand, based on the likely frequency of these events. While tsunamis funnelled into shallow shores can tower tens of metres, most tsunamis are weak with heights of only a few centimetres. Recent computer modelling at the US-based National Oceanographic and Atmospheric Administration has indicated very rare, larger tsunamis would still only stand at 3m-7m on reaching the shoreline.
What's more, regulators would have also expected Japan's sea-walls to prevent flooding at the Fukushima plant. Around 40 per cent of Japan's 22,000 mile coastline, including the coast close to Fukushima, is lined with concrete sea-walls or other structures designed to protect the country from high waves and typhoons.
As Abram says: 'They could say we'd like you to design to even higher seismic and tsunami standards... but an engineer or scientist would argue very strongly that you can get greater societal payback if you invest in something that represents a bigger risk to a greater number of people.'
In the days following the explosions, plant workers frantically injected a mixture of seawater and boric acid into the reactor vessels of reactors 1, 2 and 3, via a fire extinguisher system line in a bid to minimise fuel-rod damage. Seawater was intended to cool the reactor fuel rods while boron was to absorb neutrons and reduce radioactivity.
According to Grimston, the situation at the reactors looked 'just awful'. 'Getting the seawater into the cores is really important,' he says. 'If you can bathe the reactor in seawater and the fuel rods are melting, then at least there is some protection in this.'
The technique met with partial success. The much feared 'nuclear meltdown' did not happen but fuel rods were partly exposed. This will have led to the rods overheating and releasing highly radioactive fission products, some of which probably escaped when plant workers vented gas from the reactors to the environment, causing radiation spikes that led to plant evacuations.
'Given the magnitude of the earthquake and tsunami, then is it surprising that fuel damage was sustained?' Grimston asks. 'In an immediate sense the reactors shut down safely but even when shut down, the reactor produces decay heat due to fission products in the core. If you remove all cooling capacity from a system and continue to add heat, the temperature will go up.'
But temperatures weren't only rising in reactors 1, 2 and 3. On 15 March, the same day as the third explosion at reactor 2, a further explosion was heard at a fourth reactor. This was one of three reactors at the plant that had been shut down for routine inspections.
This explosion damaged the rooftop area of reactor 4 and part of adjacent reactor 3. The spent fuel pool inside reactor 4, located at the top of the reactor building, caught fire just a few hours later. Plant workers extinguished the fire but were later evacuated as radiation levels yet again rose.
The Fukushima Daiichi plant stores thousands of tonnes of fuel – figures from various sources range from some 1,700-3,400t – in the spent fuel pools of the six reactors as well as a joint pool. Given the cooling systems failure, cooling was lost at all seven spent fuel ponds but according to Japan's Ministry of Economy, Trade and Industry, reactor 4 had been entirely de-fuelled for maintenance in November 2010, and of the six reactors, contained the greatest number of fuel assemblies. Without being cooled the temperature in all fuel pools was rising, but TEPCO now suspected the water in the spent fuel pond at reactor 4 was boiling.
Helicopters cooling the spent fuel
Spent fuel rods are usually stored in 15m-deep pools for a few years while they become less radioactive. The water acts as a radiation shield and prevents the rods from over-heating; take it away and gamma radiation is very high while the zirconium alloy fuel rod casings will overheat and fail, releasing radioactive material.
'Reactors 5 and 6 were of less concern but reactor 4 was entirely de-fuelled with all the fuel going into its spent fuel ponds,' explains Grimston. 'Now spent fuel is serious stuff... and following the explosion, workers were getting ready to get more water into the spent fuel pond.'
Within hours, military helicopters were dispatched to pour water into these damaged reactors and spent fuel pools and fire-fighterssprayed reactors with high pressure hoses. Immediate disaster was averted and two days later TEPCO claimed the temperatures around the spent fuel ponds had dropped.
As days have turned into weeks, hundreds and hundreds of riot police, military personnel and firemen have continued to replenish water across all reactors and spent-fuel ponds, using high-pressure water cannons, fire engines and concrete pump trucks. TEPCO has also installed 1km of new cable and replacement switchgear at the plant and is replacing motors and switchboards submerged by the tsunami in the hope of restoring mains electricity to the plant to re-start the cooling systems.
But despite the massive emergency efforts, smoke still emanates from four reactors. The fuel rods are stable relative to the early days, but radiation fears are not. Elevated levels of iodine-131 have been measured in local seawater while numerous countries have detected mild levels of the same radioactive isotope and caesium-137 released from the power station in imported Japanese food.
Levels are a lot less than Chernobyl, although Japanese government has estimated the overall radiation release to the atmosphere from Fukushima to be 10 per cent that of Chernobyl and has banned food shipments from prefectures around the power station.
One question has been asked again and again: how safe were the reactors? All reactors used a General Electric light water reactor design, the boiling water reactor, with reactors 1 to 5 having the so-called Mark 1 containment system, developed in the mid-1970s. Reactor 6 had a slightly different containment design.
Myriad news articles claimed the containment to be less robust than the steel and cement 'tomb' used in the alternative light water reactor design, the pressurised water reactor (PWR). What's more, comments from US Atomic Energy Commission memos raising concerns over the safety of the GE Mark 1 containment system back in the 1970s have been repeatedly aired.
Different plants, different reactors
Abram confirms that the PWR containment is bigger, thicker and can withstand higher pressures than the BWR containment. 'As we saw in Three Mile Island, this containment worked very well and the radioactive release was negligible, with the effect on the environment and people very small indeed,' he says. 'This hasn't been the case at Fukushima but it would be unwise to draw too close a parallel to the two events as Three Mile Island wasn't hit by anything as devastating as the combined earthquake and tsunami.'
In its defence, GE states any differences in the primary containment systems used in different reactor designs are a function of different operating characteristics. For example, its containment systems use the suppression pools to reduce pressure in the containment vessel and, as such, are smaller than containment systems in a PWR, which doesn't have such pools.
Meanwhile, the energy giant also states that in the four decades since its first design, the containment system has been modified to address technology developments and changing regulatory requirements across the world which 'was not driven by threats of lawsuits from utilities' as claimed in some media reports. 'We understand that all of the BWR Mark I containment units at Fukushima Daiichi had addressed these issues and received modifications in accordance with Japanese regulatory requirements,' GE spokesperson Michael Tetuan asserts.
But, as Abrams also points out, while the GE Mark 1 boiling water reactor is a well established design, it has never been licensed in the UK. 'I don't say it is not licensable in the UK but nobody has ever brought an application to license such a design here,' he says.
While the disaster-stricken plant may have adhered to Japanese regulations, which follow the same principles as US and UK regulations, Abrams adds: 'Every single nuclear country in the world will now be looking at its nuclear regulations to see what lessons can be learned.' *