Community Coalition Against Mining Uranium (CCAMU)
Citizens’ Inquiry on the Impacts of the Uranium Cycle
Home Scope of the Inquiry Participants How to participate Counties & Municipalities
About CCAMU CCAMU Supporter Registration Location & Dates Supporting Organizations
   

Safety Concerns with MAPLE and ACR reactors not the same

Gordon Edwards

This is to clarify the safety concerns surrounding the ACRs (Advanced CANDU reactors) proposed for Northern Alberta and the MAPLE reactors built at Chalk River, Ontario.

The ACR 1000 reactors are very different from the MAPLE reactors. They do not have the SAME design flaw as the MAPLEs, but they may have a SIMILAR design flaw.

In technical language, the MAPLE reactors were designed to have a "negative power coefficient of reactivity" (PCR) but in fact they have shown the opposite behaviour -- namely a POSITIVE PCR. For the sake of safety, a negative PCR is good and a positive PCR is bad. Neither AECL nor CNSC understands WHY there is a positive PCR, and that is troubling, because it means they do not understand what is happening inside the reactor. When you don't understand what is going on, you are in danger of having some very unpleasant surprises.

The ACR reactors were designed to have a "negative void coefficient of reactivity" (VCR) but the staff of the U.S. Nuclear Regulatory Commission stated in a newsletter that they had concluded the opposite -- that under certain accident conditions, the ACR reactors could have a "substantially" POSITIVE VCR. For the sake of safety, a negative VCR is good and a positive VCR is bad. Since nobody has ever built an ACR up to now, what we have is a disagreement between two groups of experts -- one Canadian team from AECL, one American team from USNRC. Who can say who is correct? But it is troublesome that such a disagreement exists at this stage, for it suggests that there is no clear understanding of what exactly might be going on inside the core of an ACR reactor. This, coupled with the fact that AECL was WRONG about the negative PCR in the MAPLEs, raises a serious matter of concern.

At the moment the situation with the ACR design is unclear. Since no ACR reactor has been built, all the analysis is done on paper and in computer models. However when things are not clear in matters of nuclear safety, it does not bode well -- as the MAPLE experience shows.

Let me explain.

The MAPLE reactors are small -- generating about 10 megawatts of heat -- compared with an ACR 1000 -- generating about 3300 megawatts of heat. (1100 megawatts are converted into electricity, the other 2200 megawatts are "waste heat" which goes into the environment in one form or another.)

The MAPLE reactors produce no electricity. They are dedicated to the production of "isotopes". These isotopes are radioactive materials which are created through the bombardment of various "target" materials by neutrons -- fast-moving highly penetrating subatomic particles that are released during the fission of uranium atoms in the fuel of the reactor. Some of those neutrons are absorbed by atoms in the target materials, and those atoms are thereby converted into new radioactive materials. These are manufactured radio-isotopes that can be used in medicine, industry, and research.

The ACR-1000 reactors produce electricity. The fissioning of the uranium atoms in the fuel generates a tremendous number of neutrons and a great deal of heat. That heat is used to boil water in the "steam generators", and the steam is used to spin a turbine, thereby generating electricity.

In all nuclear reactors, it is important that the number of neutrons is kept within limits -- not too few, or the reaction will come to a halt; not too many, or the nuclear chain reaction will "run away" -- accelerate too rapidly for effective control. The result of a run-away accident, if not very quickly terminated (within seconds) can lead to the destruction of the core of the reactor and an unwanted release of energy which in the worst cases can lead to explosions of sufficient magnitude to breach the containment and release a cloud of radioactive gases and vapours into the atmosphere.

Because of this fundamental safety concern, the designers of nuclear reactors talk about various "coefficients of reactivity". A positive coefficient of reactivity means that you have a positive feedback mechanism which causes the population of neutrons to rapidly increase -- evidently an undesirable feature from the point of view of safety. A negative coefficient of reactivity means that you have a negative feedback mechanism which causes the population of neutrons to dwindle -- a desirable feature from a safety point of view, although it may complicate the operation of the reactor as it could lead to an undesired shutdown in the worst cases.

In the small MAPLE reactors, there is supposed to be a negative "power" coefficient of reactivity (PCR). This means that when the power of the reactor is increased, the population of neutrons should be inclined to dwindle rather than to accelerate; but low-power tests have shown that the opposite happens -- an increase in power leads to a slight acceleration in the population of neutrons. As of the summer of 2007, AECL had identified seven possible reasons why this unanticipated behaviour might be occurring, and they had eliminate two of those possibilities as likely causes. This leaves five more possibilities to investigate, and it may be that none of them are the culprits -- there may be something else going on which they have not yet identified. In the large ACR reactors, there is supposed to be a negative "void" coefficient of reactivity (VCR). This means that if a pipe breaks and so the core of the reactor loses coolant (the cooling water that flows through the core), the population of neutrons should go down, not up.

A loss-of-coolant accident (LOCA) can be very serious, leading to possible over-heating and damage of the fuel bundles, which in turn would lead to large releases of radioactive steam, gases and vapours into the plant's atmosphere. If the loss-of-coolant is prolonged, even if the reactor is completely shut down within seconds, the intense heat generated by the radio-active materials in the fuel (the fission products) can cause melting of the fuel at a temperature of about 5000 degrees Fahrenheit. Of course there are numerous backup systems designed to prevent this kind of overheating that could lead to major core damage and large-scale radioactive releases: high-pressure, medium-pressure, and low-pressure emergency core cooling systems (ECCS), and sprays within the containment building designed to condense the radioactive steam in order to prevent over-pressurization and consequent rupture of the containment building.

But if there is a POSITIVE void coefficient of reactivity, it means that whenever a loss-of-coolant accident (LOCA) occurs, the nuclear chain reaction will tend to ACCELERATE rather than slowing down, which just makes the situation worse.

Bad enough to have a LOCA, but even worse to have a potential run-away accident at the same time as the loss-of-coolant accident. Hence the desirability of having a NEGATIVE void coefficient of reactivity, which is what AECL has designed for.

Since the ACR reactors have a smaller volume than traditional CANDU reactors, and since they operate at a much higher power level than traditional CANDU reactors (more than twice as powerful as the Pickering A reactor near Toronto), and since they do not have Vacuum Buildings to suck up the radioactive gases, ACRs already have less of a margin of safety in dealing with a potential loss-of-coolant accident than the traditional Ontario-based reactors. But if there is a positive void coefficient of reactivity also, contrary to the intentions of the designers, then the ACR reactors will pose added safety problems.

Anyone wishing to know more about Core Meltdowns in CANDU reactors, or about LOCAs or ECCS, please consult:

http://ccnr.org/CANDU_Safety.html