APR: your source for nuclear news and analysis since April 16, 2010

Tuesday, July 26, 2011

Nuclear Energy in Japan: Genkai pressure vessel embrittlement

Today in Japan there are a number of reports in circulation concerning a report given by a Tokyo professor in the field of metal physics to a public watchdog group concerning supposed near-disastrous conditions at Genkai No. 1 in terms of the structural integrity of that plant's reactor pressure vessel.

The public watchdog group, Citizens Nuclear Information Center, is a Japanese non-profit that is WHOLLY ANTI-NUCLEAR. (This is flag number one.) The professor is an expert in what Japan Times says is the field of metal physics, but apparently so far as can be determined at this point has no real working knowledge of the history of reactor pressure vessel development or the study over many decades of neutron embrittlement. (This is flag number two.)

The presentation of data from this report, quotes from its author, and more are seen in the anti-nuclear hyperbole filled extravaganza linked below. If you can tolerate it, please read it and then return here for a sober discussion.


Now, for some sensible commentary.

In recent months much discussion has been given to the pressure vessels at the Fukushima Daiichi site, and whether or not they have failed. As such, then, many people are now at least partly aware of what a "pressure vessel" is, where it is located in the plant, and what its job is... which is to contain and support the reactor core, the control rods, and other associated equipment.

Many decades of research have been conducted concerning the pressure vessels in reactor plants. Since the pressure vessel is made of steel, and surrounds the reactor core, it is natural that it will suffer some sort of mechanical degradation over time due to the exposure to a very high neutron flux. Without going into minute detail, we can simply say that the neutrons will tend to collide with and displace atoms in the crystal structure of the metal of the pressure vessel, leading to local, very tiny defects. The overall effect on the metal of the pressure vessel is to force it to begin to act in a more brittle way than in a ductile way. A failure in a brittle fracture mode is sudden, with rough surfaces between the broken pieces while a failure in a ductile mode shows the effect of stretching and deformation of the metal (called "plastic deformation") before it fails.

Now, this isn't to say the metal just eventually fails. No reactor pressure vessel has EVER experienced brittle fracture. None. Zip. Zero. This is one of the Design Basis sorts of things that is absolutely intolerable, was recognized at the outset of nuclear energy's development, was constantly studied and investigated and tested. Many thousands of hours of testing, millions of man-hours and very much material have gone into embrittlement studies just for reactor pressure vessels alone.

Right here I have the first three volumes I grabbed off the shelf to begin to look at what the Tokyo professor was saying about the DBTT, or the Ductile-Brittle Transition Temperature at Genkai No. 1. This is a temperature limit which is determined at first theoretically, then modified over time partly theoretically and partly by the examination of actual pressure vessel metal samples (chunks, really) that are placed in the vessel during refuelings and later pulled out for analysis to see how they have changed. Probably a thousand samples like this are presently "cooking" inside reactor pressure vessels world wide and have been for thirty years, and have been being tested like this for that long. Anyway, the first two books are:

Reactor Structural Materials: Engineering Properties as Affected by Nuclear Reactor Service. ASTM Special Technical Pub. No. 314; published 1962

Radiation Embrittlement and Surveillance of Nuclear Reactor Pressure Vessels: An International Study. ASTM Special Technical Pub. No. 819; published 1983

As can be seen simply from the titles and publishing dates, the American Society for Testing and Materials has been involved with testing of reactor pressure vessel metals for many decades - to say nothing of all the other organizations, regulatory agencies, private companies and government laboratories here and world wide.

Further is this volume:

Toward Improved Ductility and Toughness. Published by The Iron and Steel Institute of Japan and The Japan Institute of Metals after papers delivered at Kyoto International Conference Hall, October 25-26, 1971. Volume printing courtesy Climax Molybdenum Development Co. Ltd. Japan. (Of interest in this volume is the paper presented by Reuther/Zwilsky of the US Atomic Energy Commission Division of Reactor Development and Technology which was entitled "The Effect of Neutron Irradiation on the Toughness and Ductility of Steels.")

The result of looking at the data in these three volumes is that the number given for the DBTT at Genkai No. 1 BY THE OWNER-OPERATOR is a safe number which is not out of the norm by far. The described rise in the plant's DBTT over time is also along curves presented for both A302B and A533B pressure vessel steels in use at the time that Genkai No. 1's pressure vessel would have been fabricated, so far as I can determine at this point.

Having said that, it remains unclear for the moment exactly which steel alloy this plant incorporated. I cannot yet find a complete enough version of this report, or any statement by any official agency, as to which alloy was in use. However, the given DBTT of 80C by the owner-operator is in line with graphs in these volumes and the expected rise to 91C in two further decades is certainly workable.

When persons unfamiliar with the technology dive into reports, and studies, it's very easy to misinterpret some portions. Here is a quote from the last named source report:

"It would thus appear that for the weak direction, irradiated A533B steel could fracture in a brittle mode ... in the presence of a large flaw."

This says nothing by itself of the dimension of the material, the total irradiation, the actual application of stress, the nature of the flaw itself (metallurgical, weld defect not normalized, etc.) Taken right out of context it may look alarming, but it is part of a process being used to ensure that brittle fracture doesn't happen.

It still hasn't.

This author cannot at the moment find any clue as to what the defect stated in the Genkai pressure vessel is supposed to be. Studies are done assuming a worst-case manufacturing defect, or weld defect. Whether this exists or not specifically in the Genkai vessel is not known. Kyushu Electric does say that the 80C DBTT is high, but acceptable even in the linked article.

This author did find reference in DOE documents to a study performed in Japan in 1975 wherein samples of pressure vessel steel, with welds, were tested and found to have high DBTT's. However, the report states that data from this published series of tests is incorrect because the metal was not normalized after the weld was performed, which it would have been in any reactor pressure vessel anywhere -- so in other words the test does not apply to reactor pressure vessels. It may be that some data from this test series were used -- but again until the full report by this Japanese professor appears in English, or until Kyushu refutes it publicly in English we cannot be sure exactly what the details are. What we do know are the following things:

-The article I linked is full of typical anti-nuclear press operative words. Note the reference to pouring ice water into a hot glass. Ludicrous. The press does things like that to make you think you've actually conducted a really applicable experiment in your own home before.

-The professor's material quoted in all sources that are running it gives NO sources for data either as to the RPV material or to the supposed worst-case defect or to the testing performed. This kind of thing is always suspicious.

-The agency to which this "report" was submitted probably does not care too much about actual operational fact, metallurgical research or the decades of neutron embrittlement research. All it does is print "nuclear = bad" any way it can. Submitting a "report" like this, to an "agency" like this really has no official standing whatsoever. It's good for citizens to be curious and informed, but the anti-nuclear agenda here could not be more clear.

-At the moment, what Kyushu is saying about the DBTT seems in line with available information here in print and with several documents I read this morning from the DOE and NRC about pressure vessel embrittlement - the most recent of these dated 2010. As such, considering this author's review of materials from practically forty years' worth of neutron embrittlement study there appears nothing particularly alarming about the numbers Kyushu is giving at this time.

A rising DBTT really means that the plant is limited operationally as to how fast it can cool down. Assuming the plant is operated at full or near full load all the time (called a 'base load' plant in the business) the rate of cooldown won't be affected by following system load demand. Careful operational control to limit rates of heating up the plant and cooling it down have been employed at all nuclear plants for their whole existence, even back to the first successful PWR (that would be the STR Mark I, or the S1W plant at NRTS which was started up in 1953) so that this operational aspect is absolutely nothing new, nothing novel, nothing alarming or even in most cases particularly noteworthy.

I only go into this much detail to refute this report because it tries to go into so much detail -- wrongly applied, it appears -- to make the neutron embrittlement issue look so much more important than it really is.

Naturally since this story has legs in the Japanese press and has been sent to (but oddly not published on line yet by) a big anti-nuclear group in Japan, we'll keep watching for further technical details or Kyushu or NISA press releases and will update here with whatever way this shakes out when more details emerge and are firmed up.

11:10 AM Eastern Tuesday July 26, 2011


  1. Why do pro-nuclear groups in Japan sit back and take a licking from these guys? Don't they care about accuracy? Don't they care about their jobs? Do they even have any pro-nuke blogs over there at all? And is anyone here offering them any air support in facts and historical records? For their own welfare they should nip this guy in the bud!

    James Greenidge

  2. @jimwg: It sure seems as if they don't have anyone pro-nuclear blogging there in English, anyway!

  3. > Careful operational control to limit
    > rates of heating up the plant and
    > cooling it down have been employed
    > at all nuclear plants

    How close did each of the Fukushima plants come to their limit rates for heating up and cooling down at the end? I realize each of the four was different in many ways.

    Perhaps you can from that extrapolate if each of them went through these last events but had been under a DBTT restriction, how close the event was to the limit, for how long, in which direction.

    Compare actual performance to criteria in the docs, and base hypotheticals on that.