Sunday, May 6, 2012

Overheated rods & rhetoric

A little knowledge is sometimes a dangerous thing - particularly when fundamentally incomplete technical knowledge is used to make sweeping engineering recommendations. The latest example of this is the concern over the spent fuel storage pools at Fukushima Daiichi Unit 4, which has been getting attention from several corners. First, there was U.S. Senator Ron Wyden (D-OR), a ranking member of the Senate Energy and Natural Resources committee, who recently toured the stricken Fukushima site and released a very widely reported statement that, "things were worse than reported." In particular, Wyden has singled out the spent fuel pools at Unit 4 for unique concern, calling on both the Japanese and U.S. governments to see to it that the rods are safely relocated elsewhere, citing their storage in unsound structures close to the ocean. Wyden has pushed the NRC and others to relocate these spent fuel rods to dry cask storage elsewhere.

As for Wyden's technical credentials for making this assessment? A J.D. in law and his self-assurance in a Senator's unerring technical omniscience.

I suppose it probably doesn't occur to the Senator that relocating spent fuel rods out of the damaged building is no mean feat, given that the rods which will be relocated need to maintained underwater while they are transferred into concrete casks (in this case, mostly for radiation shielding purposes) using heavy cranes. Meanwhile, TEPCO has already reinforced the damaged building, addressing the concern he has over future tsunamis further damaging the weakened building and leading to a release into the environment. Its current plans call to begin removing spent fuel for relocation within the next two years. To emphasize - this is not a problem that relevant technical experts were ignorant of until one brave Senator stepped in and decided to lead.

Of course, to be fair to Wyden, as Dan Yurman points out Wyden is clearly not the only politician suffering from an acute hubris on technical matters.

Overheated rods & rhetoric

spent fuel pool
A spent fuel pool (Image: IEEE spectrum)
At least the good Senator can be forgiven for his enthusiasm however, as it's not nearly as obnoxiously hyperbolic as certain other accounts going around the internet. Anti-nuclear activist and self-described nuclear "expert" (to use the term rather loosely) Robert Alvarez has been shopping around the dangers of spent fuel pools for some time, specifically focusing his ire upon the rods contained in the spent fuel pool at Unit 4. This of course is not a new topic for Alvarez, who has gone so far as to argue that such pools are "a ticking time bomb" and that the U.S. needs to move toward dry cask storage of all spent fuel as soon as possible. (More on why this is silly at best and potentially a dangerous misplacement of priorities in a moment.)

unit 4 spent fuel pool
Spent fuel rods at Unit 4 (Image: IAEA)
Alvarez's latest work, "Why Fukushima Is a Greater Disaster than Chernobyl and a Warning Sign for the U.S.", hits a new low in terms of outrageous hyperbole. Let's start with the headline premise - Alvarez asserts that the potential danger - a release of radioactivity from the spent fuel rods at Unit 4 - is already worse than something which actually happened - i.e., the Chernobyl disaster. (Perhaps aware of this seeming logical contradiction, Alvarez walks this back to "may be worse" in the first sentence.)

The basis of his reasoning? 1) Spent fuel contains very large amounts of radioactivity, 2) The spent fuel pools have been exposed to air (due to the hydrogen explosion at Unit 4), 3) A collapse of building containing the spent fuel pool would lead to an overheating of the rods contained at Unit 4, 4) Somehow, this would lead to a zirconium fire and release all of the radioactivity present in the rods.

Alvarez' blog post is a perfect example of the trouble one can get into when one extrapolates from a small bit of knowledge to a larger technical issue.

Taking it point-by-point - first we have this:
Several pools are now completely open to the atmosphere because the reactor buildings were demolished by explosions;
First of all, it should be noted that spent fuel pools are generally kept at room temperature and atmospheric pressure to begin with. A spent fuel pool, at its core, is essentially a very deep, very large swimming pool (which is also very radioactive as you reach the bottom). At the top, radiation levels are low enough to safely work without problems - you can even look down inside and see the eerily beautiful blue Cerenkov glow if the lights are dark enough. As for containment? The explosion at Unit 4 was in the secondary containment, which is essentially a thin* metal reinforced concrete shell - again, namely because spent fuel rods are un-pressurized and not at the kinds of temperatures found in the reactor. (In other words, the same kinds of phenomena involved in a core melt aren't relevant here.) The primary containment in any spent fuel pool is the water itself, which isn't hot enough to be going anywhere.

*(Edit: "Thin" being relative to the primary containment, which is 4-8 feet thick; most of my understanding of the secondary containment comes from diagrams such as this one, or this one via TheEngineer which bears much greater detail)

Moving on:
As more information is made available, we now know that the Fukushima Dai-Ichi site is storing 10,833 spent fuel assemblies (SNF) containing roughly 327 million curies of long-lived radioactivity About 132 million curies is cesium-137 or nearly  85 times the amount estimated to have been released at Chernobyl. 
So what does this mean? Without context - absolutely nothing. What Alvarez is trying to imply is that in the circumstance that these materials were released into the environment, the consequences would be far worse than Chernobyl. The problem? Alvarez presents no credible physical mechanism for this to happen.

Then there's this:
Also, it is not safe to keep 1,882 spent fuel assemblies containing ~57 million curies of long-lived radioactivity, including nearly 15 times more cs-137 than released at Chernobyl in the elevated pools at reactors 5, 6, and 7, which did not experience melt-downs and explosions.
Why is it not safe? Well, other than the fact that spent fuel is radioactive, Mr. Alvarez doesn't say. An industrial blast furnace is also not a safe place to be, but that certainly doesn't prevent their use. Instead, we actually take precautions to use them safely - the same way spent fuel pools use deep levels of water to both cool the fuel and shield the high levels of radioactivity.

To wit: certainly no one would want to be next to a spent fuel assembly without the shielding provided by the deep pool of water. (With this shielding, the levels of radiation are low enough where it is quite safe to stand above the pool and look down inside - something I have had the opportunity to do before). But for this radioactivity to be truly disastrous (rather than simply being a dangerous but extremely localized nuisance), something has to cause the radioactive materials in the fuel to change state - i.e., to either melt or be carried away ("lofted") by a fire.

In the beginning of his article, Alvarez eludes to the possibility of a zirconium fire, which he asserts could happen if the rods grew too hot. (Alvarez provides no further explanation or reference to credible technical resources beyond this.) Yet there are several significant problems with this theory. First, this would require the rods growing hot enough to ignite (if this is even possible - zirconium in solid form will not ignite, and its melting point is 1852° C). It second assumes that all of the radioactivity is uniformly lofted into the atmosphere; one of the main reasons for the magnitude of the Chernobyl disaster had to do with the fires in the reactor building which lofted radionuclides high into the atmosphere, where they spread across Europe. (Incidentally, this fire was also not from zirconium - it was a graphite fire from the reactor and control rod design being used.)

(Alvarez also rides his hobby-horse in inveighing against spent fuel reprocessing - a topic beyond the scope of this post but one which we've covered previously.)

A background on spent fuel

Spent fuel heat (click for larger)
Meanwhile, let's back up for a moment such that everyone understands what's going on. As we've covered on this blog before, spent fuel does still produce heat after the fission reaction shuts off. The remaining radioactive materials in the fuel, created both by fission and absorbing neutrons - are decaying. The quickest-decaying materials produce very high levels of radioactivity, and much of this energy is trapped in the fuel itself, heating it. Thus why spent fuel needs to be cooled following the reactor shutdown (which was the resulting source of problems at Units 1, 2, and 3).

Both this radioactivity and decay heat fall off dramatically with time, as the shortest-lived fission products decay away. Within 100 days, the heating rate and the radioactivity in spent fuel have dropped by a factor of 10; within 10 years, this drops to 1/100th of the original values.

Spent fuel radioactivity (click for larger)
Doing my own calculations using ORIGEN-S (a tool for nuclear licensing evaluation which is used to simulate spent fuel inventories), a typical assembly of the type found in the spent fuel pool would produce about 3-4 kW of heat after being stored around 1.5 years (and even less as it grows older) - or about 17-20 watts per pin (which themselves are over a meter long). In other words, while fuel which has just been ejected from a reactor poses a challenge in terms of cooling, it is difficult to conceive of how one gets the type of scenario Mr. Alvarez describes, in which something producing so little heat manages to cause these assemblies to melt or spontaneously catch fire.

A solution in search of a problem

Dry storage casks
Dry storage casks for spent fuel
Getting back to the main thread now - let's assume for a moment that this scenario, one already demonstrated to be of extremely questionable plausibility, is true - i.e., that there remains a real threat spent fuel pools, in which the cooling water is lost and the rods subsequently overheat and either catch fire or otherwise change state. So Alvarez's solution, to prevent these rods from overheating? Put them into thick concrete casks cooled by circulating air. Apparently, the same rods at risk of spontaneous combustion when exposed to air are fine if put into thick concrete casks. The logical inconsistency beggars belief.

Note that I am most explicitly not criticizing dry storage - in fact, dry storage casks have been demonstrated to be an effective, medium-term solution for isolating spent fuel from the environment. But to simultaneously assert a danger of spent fuel rods melting when exposed to air while simultaneously advocating to put them in thick concrete casks exposes a basic failure of physics reasoning, one which both Mr. Alvarez's employer and the ever-reliable science reporting of the Huffington Post are happy to embrace.

Alvarez and his sponsors at the liberal think tank Institute for Policy Studies are of course using this reasoning to go a step further, arguing that all spent fuel pools at U.S. reactors are at risk and thus need to be moved to dry storage. Let's just watch the errors compound...

First, let's go back to the decay heat issue. Generally speaking, spent fuel isn't suitable for moving into dry storage until it has cooled for a few years in a spent fuel pool - a general rule of thumb for dry storage is 5-10 years cooling time, although less is possible. The heat generated by 10-year old spent fuel assemblies are a hundredth of that generated by recently-ejected assemblies - in other words it would take one hundred assemblies stored for ten years to equal the contribution of one "fresh" ejected assembly.

If the reasoning here is to give greater safety margins for spent fuel pools in the event of a loss of cooling, dry storage is an extremely inefficient mechanism for doing so - namely because of the fact that the assemblies which are eligible to be moved into dry storage casks make at best a marginal contribution to the spent fuel pool heating. In other words, a large expense for very marginal gains in safety.

So here's how it breaks down: "newer" spent fuel rods are too hot to go into dry storage casks, and thus must be kept in the spent fuel pool to cool. Therefore, the integrity of the spent fuel pool must be maintained. Yet if the integrity of the spent fuel pool is maintained, there is no real safety reason (at least in terms of heat or radioactivity) to move older rods, which can be moved into dry storage. (Note: there are other reasons one may choose to do so - spent fuel pools are limited in terms of total capacity, based on a number of safety-related factors, including total heat as well as how closely the assemblies can be placed together in order to prevent assemblies from going "critical" and restarting the fission chain reaction. However, these are far less limiting circumstances.)

What we have is thus a classic case of a solution in search of a problem. Alvarez (and others, for that matter) have found a solution they like - dry storage - and have (by process of scientifically incomplete reasoning) connected this with a problem they see - the vulnerability of spent fuel in wet storage pools - and naturally put the two together. Regardless, that is, of whether that square peg will actually fit in said roundish hole - the solution is, apparently, to just keep pounding.

When well-meaning ignorance actually becomes dangerous

This is where I think Alvarez's (possibly well-meaning) concern actually becomes dangerous. Maintaining the integrity of spent fuel pools for "younger" fuel is vitally important - which is why some of the most recent changes recommended by the NRC as well as industry call for improvements such as better monitoring and instrumentation at spent fuel pools, along with other kinds of contingency plans to ensure water can be delivered to the pool in the case of a loss of coolant. Likewise, ensuring the integrity in the design of spent fuel pools indeed should be a priority.

But herein lies the problem with "experts" like Mr. Alvarez, who has no actually technical background to speak of - starting with the faulty premise that "wet storage" (i.e., spent fuel pools) can be eliminated entirely (they can't), we are then assaulted with faulty recommendations to move fuel out of these spent fuel pools at large expense and very marginal contributions to safety. Yet arguably these are resources that could be better spent on improvements to the safety of spent fuel pools - things like better instrumentation to know what is going on in said pools and improved emergency response capabilities (such as designing easier means of supplying auxiliary water to the pools). The focus on dry storage as a safety measure thus makes for a dangerous distraction which commits attention and resources away from more productive ends, thus potentially compromising safety as a whole.

Alvarez isn't the only one guilty of a single-minded focus on dry storage as a "solution" to spent fuel storage pools - all kinds of individuals (such as Senator Wyden above, and even some people I know of in real life who should know better...) have jumped all over this. The problem comes down to a simple failure to think things through - again, if spent fuel is too hot to be exposed to air, it's too hot to go inside a thick (thermally insulating) concrete cask. If it isn't too hot for dry storage (i.e., older fuel), then it isn't what is driving the safety issue at the spent fuel pool. Thus, in either case, it's a solution in search of a problem - given the fact that hotter fuel cannot be removed from the pool itself, it is more useful to focus upon the problem at hand.

The underlying pathology here - in other words, why seemingly simple-sounding solutions like this are so seductive - is because it gives the illusion of "doing something" about the (perceived) problem. In this case, this is done through a somewhat primitive technical analogy - we  have a thick concrete containment for the reactor as a safety mechanism, therefore spent fuel should similarly always be in a thick concrete containment. It simultaneously ignores where the solution is technically inappropriate ("younger," hotter fuel) and how it fails to address the root problem (i.e., keeping the spent fuel both cooled and well-shielded - which is done by ensuring the integrity of the water levels in the spent fuel pool). Fundamentally, it is an example of how not to do engineering - engaging in a top-down method of choosing a solution first and making it work to fit the problem.

Under ordinary circumstances, this leads to bad outcomes - wasted money and sub-optimal solutions (or even solutions that are simply inappropriate). In the worst-case scenario, this kind of thinking actually makes things worse, namely by committing time and resources away from evaluating actual safety improvements - and thus where well-meaning concern of outsiders who are fixed upon a particular solution without understanding the actual problem can actually do more harm than good.


  1. You said, "The explosion at Unit 4 was in the secondary containment, which is essentially a thin metal shell". It is easy when making an argument to get carried away a bit.

    This statement may be true about Unit 1 but not Unit 4. When I see rebar hanging from broken concrete, I do not think "thin metal shell". When the spent fuel pool needs to be reinforced, I do not think they are replacing a "thin metal shell". Yes, a thin metal shell was blown away but in Unit 4 there was damage to thick concrete support beams.

    I suppose that if the damage has been worse then the spent fuel pool would have fallen to the ground and the fuel would have not been covered with water. Has there been an analysis of what would happen if the fuel pool fell to ground level?

    1. @Anonymous @ 1:46: Well, here's the diagram of the Mark I containment.

      Insofar as I understand the sequence of events, the explosion at Unit 4 was in the secondary containment portion. As you see in the above, this is also where the spent fuel pool is.

      However, I'm likely wrong about it being metal - the diagrams I'd seen seem to imply that, but looking again at the photos, it looks like it's a thin reinforced concrete shell. I found this diagram of a similar unit at Oyster Creek:

      Judging by the photos again, it looks like the reactor building itself (not just the secondary containment) sustained more damage than I'd previously estimated. That being said, TEPCO is already reinforcing the building - however, I haven't seen any evidence (yet) that the spent fuel pool itself is in any real danger.

      Going back to the diagram above, I'm going to go out on a limb and say that pretty much the building itself would have to be collapsing for the spent fuel pool to fall to ground level. I don't know specifically of the accident analyses of this scenario, but I'm going to assume they have been performed. In any case, I think a large problem there would be simple mechanical failure of the fuel (again, still very bad). That being said, it's difficult to conceive of a plausible scenario where the entire contents of the spent fuel pool are transported into the environment, especially at this point in terms of fuel age.

  2. Excellent educational calm and collected article! Keep up the great work!

    James Greenidge
    Queens NY

  3. Steve, it is disappointing that people are so willing to believe theses ludicrous claims of extreme danger from spent fuel ponds. However, I think you are underestimating the public relations value of dry cask storage. After all, Japan doesn’t have a nuclear safety problem, it has nuclear confidence problem, and placebos are often the best treatment for psychosomatic disorders.

    Dry cask storage can be a win not just for politicians and publicity seekers who embrace it, but also for the utilities who must bring their reactors back on-line in order to save the Japanese economy (the public doesn’t care that we’re only talking about the oldest spent fuel). To quote “Dogbert’s Top Secret Management Handbook” (Scott Adams, 1996): “… you don’t live in an ideal world, … you are left with only two logical choices: 1) Do nothing and get fired for doing nothing. 2) Do irrational and unproductive things, and get rewarded for being a can-do manager who makes things happen.”

  4. Steve, what degree of risk do you think applies to the scenario of further earthquakes severely compromising spent fuel pool No 4 to the point were collapse occurs?

    Regardless of the risk estimate, let's say it did. What is your brief view of the consequences?

    1. Anonymous @ May 15: Off-the-cuff (i.e., *without* doing any calculations), I'd estimate we'd see the following: 1) Some mechanical failure of the fuel, which means we'd see some release of fission product gasses; some iodine, some cesium/strontium. However, I think it's a stretch to think at this point we'd see actual melting of the fuel or zirc fires. 2) It would significantly impede site recovery efforts - basically, it's a very hazardous local nuisance. Recovery work would likely have to back off for a short time while some kind of entombing occurred with the fuel from the pool in order to shield and contain it.

      So, this is why I think TEPCO's making a smart approach of it - stabilize the building now, get the debris out of the pool and begin relocating the fuel rods in the Unit 4 pool to the common fuel pool. (This will involve relocating some older rods from the common pool to dry storage to make space.) This is something I think I'll cover in my next post as time permits. Basically though, TEPCO is doing triage work - I think the concern over the mechanical stability of SFP4 is legitimate, just not the doomsday mongering, and certainly not the push to empty spent fuel pools writ large. In that sense, one does what they can do apply triage work and then relocate the rods to a more stable pool.

  5. (The view that collapse might occur appears to be the motivation behind the urge to move the fuel rods from the pool in the minds of many people watching. I think.)

  6. Alvarez's argument is so spectacularly illogical I have to wonder if he's receiving any sort of support from a dry storage vendor/lobbying group; though I suppose one should never attribute to malice what can be explained equally well by ignorance.

    Thank you for these well-considered analyses Dr. Skutnik.

  7. In the presence of steam above about 900degC, Zr will oxidize very violently indeed.