CANDU in Japan: different, proven, and symbiotic with the current fleet

The new Japanese prime minister, Shinzo Abe, recently told a television interviewer that he wants to see new nuclear reactors built in Japan. He said that these should be “totally different” from the current ones, which are light water designs of the pressurized-water or boiling-water variety. He should consider the Canadian CANDU. It is totally proven: CANDUs are providing, right this minute, all of the output in the “nuclear” category in Figures 1 and 2 to the left. As for the CANDU’s ability to handle possible accident scenarios in an earthquake zone such as Japan, here is a video by the Canadian Nuclear Safety Commission that describes how a CANDU station would handle an event such as the one that brought down the Fukushima Daiichi plant in northeastern Japan in March 2011.

The CNSC video does not mention a fundamental, critical safety design difference between the CANDU and its light water cousins. Light water designs all feature a single heavy reactor vessel in which all the fuel, arranged in a single large assembly made up of many individual rods configured vertically, sits immersed in ordinary water (H2O). The CANDU by contrast features a horizontal configuration of many individual pressure tubes in which many smaller assemblies, consisting also of many individual fuel rods, are inserted, surrounded also by water but in this case heavy water (D2O).

In the event of a loss of coolant accident (LOCA) and a meltdown of the fuel, there is tremendous pressure inside whatever pressurized container holds the fuel. When the water level drops below the top of the fuel, whether the fuel is arranged vertically as in most LWRs or horizontally as in the CANDU, steam forms and fills the volume that was once filled with liquid. This creates huge pressure.

Such a scenario is at the bottom of the fears of nuclear meltdown. The fears centre on the possibility that if the pressure becomes extreme enough, the pressurized vessel holding the fuel will explode as it ruptures, violently ejecting radioactive fission products into the surrounding area.

Light water designs defend against this possibility by employing brute force. Pressure vessels are made of steel so thick that they can, it is hoped, withstand extreme pressure. The Fukushima-Daiichi event of March 2011 demonstrated that the hope was justified. Three 1970 vintage reactors melted down, and not one of the pressure vessels failed.

The CANDU designers defended against this highly unlikely eventuality not with brute force, but with guile. The consequences of the failure of a pressure vessel become more extreme as the vessel gets bigger. So instead of a single large pressure vessel as in an LWR, the CANDU features many, much smaller, pressure tubes—the CANDU 6 reactor has 380 of them. In the unlikely event of a large LOCA, where there is simply no coolant to prevent the fuel from overheating and melting, the CANDU’s fuel channels would over-pressurize and explode into the surrounding heavy water moderator. The calandria—the large horizontal cylinder that holds the pressure tubes and moderator—would then sag to the concrete floor of the reactor building, at which point the controlled depressurization of the reactor or vacuum building would take place as depicted in the CNSC video. There would, and could, be no failure of the containment. No fission products would be ejected uncontrolled into the surrounding area.

There is simply no possibility of the violent ejection of the reactor core, because that opportunity has been designed out of the equation.

The Japanese prime minister’s stated preference for new reactor designs reflects his desire to effect a political compromise between his need to assuage public fear of a meltdown and Japan’s urgent need for nuclear-generated electricity. The public must be mollified, even though the Fukushima Daiichi meltdowns have, after 1999 days, not produced a single human casualty.

It is not known what the PM meant when he said the new reactors would be “totally different” from the current ones. He might be referring to fast-neutron reactors cooled with liquid sodium, of the type Japan has been developing in collaboration with France and the U.S. That would be a huge step forward in the world energy scene.

But Abe has more pressing nuclear needs, beyond getting Japan’s 45 currently shut-down reactors back up and running. His electricity-starved and environmentally responsible country needs more than 47 reactors (there are two currently running). He needs a reactor design that is solidly proven, and that could dovetail with the current light water fuel cycle. This rules out every single current design on offer in the world, except the CANDU.

The CANDU would also offer the more or less immediate ability to begin addressing any used fuel issues that might also be plaguing public re-acceptance of nuclear energy. The CANDU, through the proven DUPIC fuel cycle (DUPIC stands for Direct Use of Pressurized Water Reactor Spent Fuel in CANDU), could burn used fuel from Japan’s PWRs.

Suddenly, Japan seems on the verge of becoming the world’s leading nuclear nation. It is adept in fuel recycling already, and has been at the forefront of fast reactor development, which would represent an elegant solution to the long-lived heavy isotopes in used fuel. And by adding the totally proven and inherently safe CANDU to their technological repertoire, Japan would finally achieve true flexibility in all their nuclear options.

11 comments for “CANDU in Japan: different, proven, and symbiotic with the current fleet

  1. seth
    January 2, 2013 at 16:03

    This is funny. The idea that a modern Japan would look at an ancient Candu rather than the Toshiba AP-1000.

    If a PHWR was what was needed they could buy more modern PWHR’s from India or India’s new ACR-1000 based future model. Unlike the Candu which is trying to make money pushing last centuries technology they got for free, India is actually continuing development of the PHWR.

    If they really wanted Canadian technology they would be much better off buying David leBlanc’s DMSR. They could have those things rolling off the factory production line in 5 years.

    • January 3, 2013 at 08:06

      Seth, I had to smile. “Ancient”? Sure, the technology was developed in the last millennium but so was the Otto cycle internal combustion engine. Lexus, the creme de la creme of high end luxury cars, still builds most (actually, I think all) of its cars around Otto-cycle ICEs.

      The modern CANDU is a pretty sophisticated machine. The one currently on offer, the EC6, is an impressive beast — here’s a rundown from CANDU Energy.

      I agree totally on David LeBlanc’s OMSR — it is very interesting and deserves support.

    • Heather
      January 3, 2013 at 08:16

      seth, the new CANDU reactor designs are quite leading edge. Why just look at existing installations? Please research before (mis) writing, and learn how to write properly.

      • seth
        January 3, 2013 at 21:17

        My dear Helen.

        I sorry you haven’t been following the history of the world nuclear development. Perhaps your reading skills match your perception of my writing ones.

        Gen III reactors are obsolete now with few ever been built. The Candu which history tells us despite dubious claims to the contrary requires an extremely iffy core replacement 25 years in the future would have to be the least likely of these sales if it weren’t for Ontario government politics. Given recent events, it is unlikely the Candu’s parent company will survive its godfather in the current government much less 25 years into the future in time to start replacing tubing in any current/near future sales.

        Currently only Gen III+ reactors are being proposed outside of Ontario.

        The experts at AECL recognized this and to be competitive had more or less completed the design of the best in world Gen III+ ACR-1000 unit. When Ontario under the influence of a rabid green element rejected the AECL proposal, the company was effectively doomed with the current extreme far out ultra right wing government, philosophically opposed to all government business activity in Canada other than it seems that run by Chinese military owned state enterprises. Because of the close ties between the government and a prominent engineering firm known for its close ties to the Gaddafi government in Libya, AECL was gifted to its cronies with a present of $60M to document the 6E design.

        The new owner as part of its gift, pledged not to do any development work on the 6E beyond that $60M and not to continue work on the ACR-1000. A utility would be foolish indeed to make what could be a 60 year relationship with a nuclear reactor when its parent company has made such a vow.

        The Indian version of the Candu on the other hand is planned to be the workhorse of the India nuclear fleet and is the recipient of significant development funding. Rumor has it that the new owner actually sold the ACR-1000 development work to India, so I would expect this unbeatable successor to the 6E will be trumping all 6E sales shortly.

        Nope a utility would be foolish indeed to commit to a one time purchase of a doomed obsolete product.

  2. Jagdish
    January 3, 2013 at 20:12

    Japan has a lot of civil plutonium assets, next only to the UK. While the UK has forgotten its value, the Japan could use it as Thorium-Plutonium fuel in its reactors including the horizontal PHWR’s if they get there.
    Japan, provider of heavy engineering components to the world’s reactors going to multiple lighter tubes! Not even India or Canada have given it a serious thought for other than the PHWR’s yet.

  3. D. Burjorjee
    January 8, 2013 at 14:57

    I find the negative comments about the CANDU strange. AECL/Candu Energy have done studies in conjunction with the Chinese in burning thorium in Candus, in burning extracted enriched uranium from LWR used fuel with tailings from the enrichment prrocess, in a CANDU. Surely these options are far preferable to using liquid sodium which so far has been a nightmare in every application.
    The main drawback to the CANDU is its need for re-tubing, and I understand that metallurgical advances have significantly increased the life of the pressure tubes.
    The low energy density of the CANDU core does make it more costly, but certainly adds to safety margin in the event of an accident. It is much more economical of uranium than LWRs or any enriched reactor; LWRs enrich by a factor of 5 to increase burnup by a factor of about 3. This means that sigificantly moore material is mined with the accompanying environmental destruction.

  4. January 11, 2013 at 10:35

    (hoping this site handles HTML properly–I HATE not having a preview!)

    In the unlikely event of a large LOCA, where there is simply no coolant to prevent the fuel from overheating and melting, the CANDU’s fuel channels would over-pressurize and explode into the surrounding heavy water moderator.

    I’d heard from someone years ago that the CANDU moderator tank could be dumped, which would instantly stop any chain reaction and also eliminate any reservoir of material for steam explosions or to dissolve and mobilize fission products.

    The point about burning used PWR fuel is a good one, but the supply would only last a decade or two with a full build-out.  The idea of using reclaimed Pu as the starting fissionable load for thorium fuel rods, then using the thorium rods as LWR fuel once enough fuel was bred, both eliminates the Pu and zeroes out the need for uranium imports; rods too “burnt” to be LWR fuel any more could go back to the HWR to squeeze the last dregs out energy out of them before disposal or reprocessing.  100 GW(e) of plants running at 30% thermal efficiency is 333 GW(th); at 100 GW-d/t burnup, the total burn would be about 3.3 tons/day or a paltry 1200 tons/year.

    100 GW(e) would generate most of Japan’s annual electrical consumption (~105 GW(e), source) if I’m reading the category definitions correctly.

  5. Udhishtir
    January 14, 2013 at 22:10

    Is “explode” the best expression to use while describing a Station Black-out scenario, particularly in the context of the postulated over-pressurization? I think this e-word is better avoided since it is likely to be misunderstood by general public (in the context of a nuclear reactor, in confusion with a nuclear bomb).

    Assuming that the pressure tubes suffer a total, full cross-sectional area breakage, — actually, I feel this is not likely since designers would have provided more than adequate capacity of passively actuated pressure relief valves in the fuel-cooling circuit — won’t the Calandria Tubes which surround the pressure tubes be of any help in providing a pathway out for the steam being formed, and there by help to reduce the pressure build up? If at all an explosion is to take place, it probably be would be due to Hydrogen formation resulting from Zirconium-Steam reactions taking place at elevated temperatures. The CNSC video says CANDU design covers this aspect.

    A main difference between the LWRs at Fukushima and CANDUs stems from the use of Natural Uranium in the latter. The energy density in the core is much less. In other words, for the same reactor size in terms of MWe output, in the event of total loss of electric power, the effort needed to cool the fuel after the reactor is shut down (tripped), using passive means, would be less in CANDU. In this respect, I would tend to believe that Natural Uranium reactors are more “fault-tolerant” than reactors using enriched uranium. This is a major advantage for CANDU.

    • January 16, 2013 at 00:41

      A reactor with the same average power is going to have roughly the same inventory of fission products and thus decay heat after shutdown.  That’s trivial physics.

      • June 11, 2013 at 20:37

        Late to this party, but … ‘Udhishtir’ speaks of energy density, but I think he means power density. The low neutron capture of heavy water means there can be more of it per watt, and this is indeed a heat-sink advantage.

        I’d heard from someone years ago that the CANDU moderator tank could be dumped, which would instantly stop any chain reaction and also eliminate any reservoir of material for steam explosions or to dissolve and mobilize fission products.

        But you don’t want to do that, because it’s at only 80°C, kept that cool by its own cooling circuit, independent of the power conversion one, and has a lot of heat capacity.

        Rather than dump it, what they have in modern CANDUs, as a backup method of turning fission off, is a gadolinium nitrate injection system. The injection is easy because the moderator is unpressurized.

Leave a Reply

Your email address will not be published. Required fields are marked *