Former Japan prime minister,Yoshihide Suga, is handed a sample of contaminated water at the Fukushima Daiichi nuclear site. (Photo: 内閣官房内閣広報室/Wikimedia Commons)
At the present time, over a million tonnes of tritium-contaminated water are being held in about a thousand tanks at the site of the Fukushima Daichi nuclear power station in Japan. This is being added to at the rate of ~300 tonnes a day from the water being pumped to keep cool the melted nuclear fuels from the three destroyed reactors at Fukushima. Therefore new tanks are having to be built each week to cope with the influx.
These problems constitute a sharp reminder to the world’s media that the nuclear disaster at Fukushima did not end in 2011 and is continuing with no end in sight.
Recently TEPCO / Japanese Government have been proposing to dilute, then dump, some or all of these tritium-contaminated waters from Fukushima into the sea off the coast of Japan. This has been opposed by Japanese fishermen and environment groups.
There has been quite a media debate, especially in Japan, about the merits and demerits of dumping tritium into the sea.
Many opinions have been voiced in the debate: most are either incorrect or uninformed or both. This post aims to rectify matters and put the discussion on a more sound technical basis.
1) TEPCO / Japanese Government have argued that, as tritium is naturally-occurring, it is OK to discharge more of it. This argument is partly correct but misleading. It is true that tritium is created in the stratosphere by cosmic ray bombardment, but the argument that, because it exists naturally, it’s OK to dump more is false. For example, dioxins, furans and ozone are all highly toxic and occur naturally, but dumping more of them into the environment would be regarded as anti-social and to be avoided.
2) TEPCO / Japanese Government have argued that it is safe to dump tritium because it already exists in the sea. Yes, tritium is there but at low concentrations of a few becquerels per litre (Bq/l). But the tritium concentrations in the holding tanks at Fukushima are typically about a megabecquerel per litre (MBq/l). In layman’s terms, that’s about a million times more concentrated.
3) TEPCO / Japanese Government have argued coastal nuclear plants routinely dump water that contains tritium into the ocean. Yes, this does (regrettably) occur as their cooling waters become tritiated during their transits of reactor cooling circuits. But two wrongs do not make a right. Moreover, the annual amounts are small compared with what is being proposed at Fukushima. A one GW(e) BWR reactor typically releases about a terabecquerel (trillion Bq) of tritium to sea annually. But Fukushima’s tanks hold about one petabecquerel (PBq or a thousand trillion Bq) of tritium – that is, a thousand times more. A much bigger problem.
4) Readers may well ask where is all this tritium coming from? Most (or maybe all) the tritium will come from the concrete structures of the ruined Fukushima reactor buildings. After ~40 years’ operation they are extremely contaminated with tritium. (Recall that tritium is both an activation product and a tertiary fission product of nuclear fission.) And, yes, this is the case for all decommissioned (and by corollary, existing) reactors: their concrete structures are all highly contaminated with tritium. The older the station, the more contaminated it is. In my view, this problem constitutes an argument for not building more nuclear power stations: at the end of their lives, all reactor hulks will remain radioactive for over 100 years.
5) What about other radioactive contaminants? Reports are emerging that the tank waters also remain contaminated with other nuclides such as caesium-137 and especially strontium-90. This is due to the poor performance of Hitachi’s Advanced Liquid Processing System (ALPS). Their concentrations are much lower than the tritium concentrations but they are still unacceptably high.
For example, on 16 October 2018, the UK Daily Telegraph stated:
“Tokyo Electric Power Co (Tepco) which runs the plant, has until recently claimed that the only significant contaminant in the water is safe levels of tritium, which can be found in small amounts in drinking water, but is dangerous in large amounts. The [Japanese] government has promised that all other radioactive material [apart from tritium] is being reduced to “non-detect” levels by the sophisticated (ALPS).
“However documents provided to The Telegraph by a source in the Japanese government suggest that the ALPS has consistently failed to eliminate a cocktail of other radioactive elements, including iodine, ruthenium, rhodium, antimony, tellurium, cobalt and strontium.
“That adds to reports of a study by the regional Kahoko Shinpo newspaper which it said confirmed that levels of iodine-129 and ruthenium-106 exceeded acceptable levels in 45 samples out of 84 in 2017. Iodine 129 has a half-life of 15.7 million years and can cause cancer of the thyroid; ruthenium 106 is produced by nuclear fission and high doses can be toxic and carcinogenic when ingested.
“In late September 2017, TEPCO was forced to admit that around 80 per cent of the water stored at the Fukushima site still contains radioactive substances above legal levels after the Ministry of Economy, Trade and Industry held public hearings in Tokyo and Fukushima at which local residents and fishermen protested against the plans. It admitted that levels of strontium 90, for example, are more than 100 times above legally permitted levels in 65,000 tons of water that has been through the ALPS cleansing system and are 20,000 times above levels set by the government in several storage tanks at the site.”
So what is to be done?
First of all, the ALPS system has to be drastically improved. After that, some observers have argued that, ideally, the tritium should be separated out of the tank waters. Some isotopic tritium removal technologies have been proposed, for example by the International Atomic Energy Agency, but the picture is complicated. The only operating facility I’m aware of, is located at Darlington near Toronto in Canada, though secret military separation facilities may exist in the US or France.
However the Darlington facility was extremely difficult and expensive to construct (~12 years to build and to get working properly), and its operation consumes large amounts of electricity obtained from the Darlington nuclear power station nearby. Its raison d’ȇtre is to recover very expensive deuterium for Canadian heavy water reactors.
Other proposed remedies will probably be more expensive. One problem is basic physics. The tritium is in the form of tritiated water, which is effectively the same as water itself, so that chemical separation or filtration methods simply do not work.
Another problem is inefficiency: with isotope separation, one would have to put the source hydrogen through thousands of times to get even small amounts of separated non-radioactive hydrogen. A third problem is that hydrogen, as the smallest element, is notoriously difficult to contain, so that gaseous tritium emissions would be very large each year.
None of these technologies is recommended as a solution for Japan: any such facility would release large amounts of tritium gas and tritiated water vapor to air each year, as occurs at Darlington. Tritium gas is quickly converted to tritiated water vapor in the environment. The inhalation of tritiated water vapor from any mooted Japanese facility would likely result in higher collective doses than the ingestion of tritiated sea food, were the tritium to be dumped in the sea.
I recommend neither of these proposed solutions.
There are no easy answers here. Barring a miraculous technical discovery which is unlikely, I think TEPCO/Japanese Government will have to buy more land and keep on building more holding tanks to allow for tritium decay to take place. Ten half-lives for tritium is 123 years: that’s how long these tanks will have to last – at least.
This will allow time not only for tritium to decay, but also for politicians to reflect on the wisdom of their support for nuclear power.
This article is republished from Dr. Ian Fairlie’s blog of September 18, 2019 and updated by him in May 2022.