There's been a significant amount of talk concerning the MAPLE CANDU reactor disaster going on in the media, both internal media in the nuclear community and external media like the National Post.
In the discussion there are at least two sides, one side, represented by folks like Harold Smith (a former AECL official who was involved in the MAPLE project) who was quoted in the National Post saying that "The MAPLE reactor operated like a dream and was/is fully capable of meeting all objectives."
The other side, represented by folks like Hugh MacDiarmid, AECL's CEO, which says emphatically that the reactors "had technical problems that had defied solution up to that point. When we looked at the possibilities, all of them were highly risky, expensive and lengthy .... We could not develop the requisite confidence level.”
Naturally both sides can't be right. This is a question of science and technology and there are no ambiguities in science. Either the reactor works or it don't. Right?
The problem is that the MAPLE reactors do work, just not as they were predicted to work. THe MAPLE reactors have what is called a positive power reactivity coefficient and to understand why this is a problem, you first have to know a few details about nuclear reactor physics.
Nuclear reactors are generally designed to have a negative power reactivity coefficient, which means that as the power increases, the reactivity inside the core should decrease. As reactivity decreases, so does the power. A negative power reactivity coefficient is an important safety feature of nuclear reactors, meant to help control the power level.
A positive power reactivity coefficient means that as the power increases, so does the reactivity, and as the reactivity increases, so does the power. That means if there is an accident the reactor power would increase rapidly until the safety systems kicked in and shut it down. The power reactivity coefficient is unrelated to the emergency reactor shutdown systems and so in the event of an accident, they would still function properly to shut down the reactor.
So it seems at this point like the question is more of a scholarly matter than one that should trip up AECL so much that the reactors must be shut down. But the regulator (the Canadian Nuclear Safety Commission) wanted to know why the reactor was not functioning as it was supposed to. Not an unreasonable request considering that you are dealing with a nuclear reactor. But AECL cannot answer it.
One might think, 'oh well, just build a whole new core and stick it in the building' which is an option, but if you don't know what went wrong the first time you built the core how can you be sure it won't happen again? Moreover, as I understand it, AECL is not allowed to experiment with the reactor on now that they know about this problem. But really, they cannot figure out what is wrong unless they run the reactors. A catch-22.
Which leads to the 'nuclear' option (pardon the pun) for the government. And there are two, neither of them pleasant. Either shut the MAPLE reactors down and stop playing with them or override the regulator and tell AECL to start up the reactors, consequences be damned.
In the former case, Canada loses its hundreds of millions of dollars investment in the MAPLE reactors and AECL's reputation is left in shambles. In the latter case, Canada's reputation on nuclear safety becomes a radioative wasteland and there will likely be reprucussions from WANO (World Association of Nuclear Operators) and the CNSC will be made toothless.
Neither option is pleasant, but Harper's decision to stop funding for MAPLE, whether it was intended this way or not, was better for the long term health of nuclear safety in Canada.