NRU shows the need to replace legalist focus with technical focus

During the last three months, there has been a significant controversy in Canada about the National Research Universal (NRU) – a fifty year old 135 MWth nuclear research reactor that is a primary source of several important medical isotopes. The story is a complicated one that deserves some thoughtful investigation and understanding. It is also one that illuminates major lessons worth learning. I have withheld comment on the issue while gathering information and thoughts suitable for sharing.

If you are new to the story, a good place to start is on one of my long time favorite sources of information about Canadian nuclear technology – The Canadian Nuclear FAQ. That site has been Dr. Jeremy Whitlock’s labor of love since the mid 1990s. Jeremy is a physicist with a lifelong interest in learning and teaching. He really knows his stuff. His article about the technical details of the controversy is titled Why was a Chalk River reactor shut down in November 2007, causing a shortage in medical radioisotopes? and it comes complete with drawings that provide a good understanding of the system design and operation.

The basic story line is that the Canadian Nuclear Safety Commission determined that the NRU was in violation of a license requirement. The regulator pressured AECL, the reactor operator, to extend a maintenance shutdown while installing the connections from a seismically qualified Emergency Power Source (EPS) to two out of eight installed coolant circulating pumps. This modification to the fifty year old design was required to meet the CNSC’s interpretation of the rules in place when the plant’s operating license was extended in July of 2006. Apparently, AECL has publicly agreed that the modifications should be made, but disputes whether or not the upgrade was a strict requirement of its operating license extension. By most accounts, the CNSC refused to approve a short term licensing amendment allowing the plant to return to operation while preparations and installation of the modified power supply were completed.

For power reactors, this type of regulatory response is accepted practice and there is never much outcry. The plant owners do whatever is necessary to satisfy the regulator. While their expensive, but low operating cost reactors remain shutdown to comply with whatever rule interpretation is current for their regulator, they purchase replacement electricity from some other source. The source is invariably a more expensive fossil fuel burning power plant. Sometimes the replacement fuel is relatively clean, but expensive natural gas, but often it is coal burned in a generally uncompetitive, dirty, inefficient old plant that is pressed back into service to meet the demand.

There is a large cost to such a scheme, but there is not really anyone who feels much pain. Most nuclear operators in the world are part of monopoly electricity suppliers (either private or government run) so they simply pass any costs on to their large customer base in the form of slightly higher electrical power rates. The contractors fixing the problem earn revenues, the competitive sources of power earn new revenues, the regulators get congratulated by anti-nuclear advocacy groups, and all seems well with the world. The extra pollution caused by replacing fission with fossil fuel combustion gets lost in the massive quantities of that material produced every day anyway.

The NRU, however, is not a power reactor – it is a research reactor whose uses include production of Mo-99 which is the parent isotope for Tc-99m. Tc-99m is used in about 70% of all radioisotope diagnostic procedures. The worldwide market for Mo-99 is not very large; the material supplied by the NRU represents between half and two thirds of the world’s supply. Not only is there a tiny supplier base, but the material also has a rather short shelf life and cannot be stockpiled in the event of a disruption in the supply chain.

As the shutdown to install the modification required by CNSC stretched from days to weeks, hundreds of medical providers who serve thousands of patients each day started to get nervous about their ability to continue performing their potentially life-saving procedures. Though all operating reactors naturally contain a certain amount of Mo-99 – it is a fairly common fission product – there are only a small number of reactors that are configured to allow access to fuel material for routine processing and recovery of individual isotopes. There were some back-up plans being developed, but all of them would take a long time to implement. Not only would there be a huge cost, but there was every likelihood that some medical procedures would have to be skipped as the supply of Mo-99 fell to the point where there was an unrecoverable shortage of Tc-99m.

Eventually, the Canadian Parliament passed a law inserting language allowing the NRU to return to service with one pump connected to the super secure EPS (the third back-up power supply) and a plan in place to finish the connection for the other pump within 120 days.

Linda Keen, the President of the CNSC had insisted that her organization’s mandate did not allow it to consider any risks from NOT operating a reactor, her interpretation of the CNSC mandate is that they are responsible for controlling and minimizing any potential for harm to the public caused by operation of nuclear facilities or possession of nuclear materials. In her mind, the proper body for balancing the risk of NOT supplying medical isotopes against the possible risk from operating without 100% compliance was the Parliament. There were also some harsh words exchanged between her and other ministers (primarily Prime Minister Steven Harper and Natural Resources Minister Gary Lunn) who had attempted to press the issue before Parliament got involved.

The story has not yet fully played out. At the end of January, Ms. Keen testified before the House of Commons Standing Committee on Natural Resources and made several inflammatory and technically incorrect statements.

In case you do not want to or cannot view the embedded video, the headline summarized from her stated testimony was “Linda Keen testified that the risk of restarting the Chalk River reactor was 1000 times higher than accepted international standards.”

The first inaccuracy is the idea that the chances of fuel failure without the back-up power supplies hooked up is “one in a thousand”. The real fact is that the assumed frequency of occurrence for an earthquake assumed to be large enough to cause a disruption in the power supplies to the existing pumps is one every thousand years. There are no records of any such quake occurring in the NRU’s location. Since four out of eight of the existing pumps have back-up power supplies that include both diesel generators and batteries, many observers believe that at least one of the pumps would remain operational even without the EPS connected. In addition, most nuclear operators would also state that they would figure out another way to provide some cooling water to the plant if there really was an extended loss of power.

The second part of the statement that is incorrect is that there is “an international standard” that requires that the chance of fuel failure be “one in a million”. The fact is that fuel failures do happen, but they do not necessarily pose a threat to public health and safety. There are still numerous barriers that keep the radioactive material away from the general public. In the case of the NRU, the very pessimistic safety analysis done for the plant licensin
g computes that a complete fuel failure of the kind that could result from a total loss of cooling MIGHT cause the most exposed member of the public to receive a dose of 2.4 mSv (240 mrem). That is roughly equivalent to the average annual dose received by a member of the public in the US, if you do not include medical procedures.

My bottom line is that regulators need to have a more complete technical understanding with more qualified staff work. In addition, any laws that are in place that forces them to make decisions about nuclear plant operations in isolation of their wider effects should be revised. Failing to operate a safe reactor in order to comply with a legal requirement may cause more negative public health and safety effects than NOT operating the reactor. In a world with a growing population, tight energy supplies and a growing understanding of the effects of fossil fuel pollution, that statement applies to power reactors as well as reactors that produce short lived medical isotopes.

Additional reading:
John Waddington . Never a real nuclear danger Op-Ed piece published February 7, 2008 by ottawacitizen.com
AECL Clarifies Inaccurate Statements by Former CNSC CEO Linda Keen
Linda Keen’s Testimony blog entry on Runesmith’s Canadian Content
AECL says nuclear watchdog misled public on reactor shutdown January 15, 2008 on NationalPost.com

This one is not actually related to the current situation for isotope production, but I came across it and thought it provided an interesting historical perspective. It talks about a time in the mid 1980s when the supply of Mo-99 was threatened by an interruption in the supply of U-235 from Oak Ridge. U-235 (also known as highly enriched uranium) is normally the target material for medical isotope production because it results in easier to process materials.
Enriched Uranium Embargo Risked Mo-99/Tc-99m Supply Journal of Nuclear Medicine.