Tracking highly enriched uranium and plutonium, the key nuclear weapon materials

U.S. plutonium disposition program: Uncertainties of the MOX route

| 11 Comments | 0 TrackBacks

The FY2012 budget request for the National Nuclear Security Administration shows that disposition of weapons-grade plutonium as MOX fuel in power reactors remains one of the largest items in the Defense Nuclear Nonproliferation program. NNSA asked for $645,721,000 for the Fissile Materials Disposition related activities (60 percent of this money - $385,172,000 - would go to the MOX Fuel Fabrication Facility at the Savannah River Site). The FY2012 request is about 10 percent smaller than the previous year request for these activities, but the MOX related expenses would still account for a quarter of the $2.5 billion ($2,549,492) that NNSA requested for its Defense Nuclear Nonproliferation program.

The direction of the U.S. plutonium disposition program has raised a number of security concerns. In addition to absorbing a large part of the funds allocated for nuclear nonproliferation, the program is likely to increase proliferation risks by supporting development of the plutonium economy. In FY2010, NNSA has reached an agreement with the Tennessee Valley Authority (TVA) "to evaluate the irradiation of MOX Fuel in up to 5 TVA reactors" (NNSA FY2012 Budget Request, p. 379). It has also been encouraging other utilities to enter the MOX irradiation program. This effort was partly a response to the decision of Duke Energy to abandon testing of MOX assemblies in its Catawba Unit 1 reactor and pull out of the program.

DOE documents obtained by Friends of the Earth in February 2011 show that Energy Northwest is "formally evaluating the potential use of MOX fuel" in its Columbia Generating Station reactor. According to Tom Clements of Friends of the Earth, the plan developed by Energy Northwest and DoE calls for irradiation of 10 to 20 fuel pins fabricated by Pacific Northwest National Laboratory would begin in 2013 or 2015. This would be followed by the use of up to eight "lead use assemblies" (LUAs) around 2019. NNSA expects that MFFF will produce the first eight fuel assemblies "a part of the facility hot start-up plan" (Budget Request, p. 400). Clements estimates that the assemblies would have to be tested in the reactor for three or more two-year fuel irradiation cycles, so the use of MOX fuel would not begin until at least 2025. Use of MOX fuel assemblies in TVA Browns Ferry BWRs would also begin no earlier than 2025, assuming NRC licenses for use of the material are secured. The FY2012 budget request apparently to include funds for these activities - the irradiation part of the plutonium disposition program includes money for "qualification of MOX fuel designs for pressurized water reactors and boiling water reactors from multiple fuel suppliers, and execution of fuel supply agreements with TVA and potentially other utilities" (Budget Request, p. 380).

Given the time required to license the fuel, there is a real possibility that the $5 billion MOX plant could sit idle during multi-year irradiation tests of MOX in BWRs. The two PWR reactors at TVA's Sequoyah nuclear plant might not be able to use all fuel produced by MFFF and further NRC-licensed testing of "lead test assemblies" will also be needed to certify MOX use in PWRs. The Sequoyah reactors also might participate in the DoE Tritium Readiness program, which may complicate the use of MOX fuel in these reactors. In January 2011 TVA and Areva signed a letter of intent regarding potential use of MOX produced at MFFF in TVA reactors. The letter, obtained by Friends of the Earth, shows that TVA has not yet made a strong commitment to MOX and that little progress has been made towards NRC-licensing of MOX use by TVA.

Among the factors that might contribute to the reluctance of U.S. utilities to enter into the plutonium disposition program are concerns about reliability of MOX fuel supply.

(with contribution from Tom Clements)

No TrackBacks

TrackBack URL:


What is so wrong with a plutonium economy? The hydrocarbon economy that we have been pursuing for the past 75 years has not provided a lot of stability, though it has enabled some pretty amazing technical achievements and comfortable living.

What is wrong with recognizing that a widely available material has 2 million times as much useful energy per unit mass and that it could enable humans to live far more comfortable and clean lives while consuming far less material every year?

The only people who should be afraid of a plutonium, uranium and thorium based economy are those who make their living by selling coal, oil and natural gas or by supplying those industries.

What's wrong with a plutonium economy? What's wrong with increasing that particular material's abundance and accessibility? What problem do I have with commoditizing an isolated fissile material stream? Aside from the ludicrous costs? None whatsoever.

Energy density tunnel vision... Let me guess, Rod, you're an engineer.

Rod Adams: There are at least two problems with the plutonium economy. First, from the nuclear non-proliferation point of view, the spread of the technology is highly problematic. It might be possible to manage these risks with proper institutional arrangements, but right now the international system is not quite up to the task. Second, the plutonium route is yet to demonstrate that it is economically viable - so far it has been supported largely by public subsidies.

The first step in economic viability is permission to pursue. If you have a religious opposition to nuclear energy or to the use of plutonium - fine. However, please do not wrap it up in an economic argument because the cost of any technology can be driven almost to infinity with the wrong kind of rule structure.

Right now, commercial nuclear fuel, with all of the overhead associated included, provides heat for US utilities at a cost of just 57 cents per million BTU. Using it produces less than 10 grams of CO2 per kilowatt-hour of electricity. Cheap natural gas costs nearly $4.00 per million BTU and burning it produces nearly 600 grams of CO2 per kilowatt hour of electricity. Burning petroleum costs more than $18.00 per million BTU at today's prices and burning it produces about 800 grams of CO2 per kilowatt hour of electricity.

Nick accuses me of the apparently horrible - to him - trait of being an engineer. Though I have been assigned that role before, I was a humanities major in college and have also run a couple of businesses.

Rod Adams: You seem to compare only the cost of fuel - then, of course, nuclear will have the advantage since most of the cost of nuclear energy is capital cost. The MIT Study estimated that fuel cost was $0.67 and $7.00 per mmBTU for nuclear and natural gas respectively, but the cost of electricity (which takes into account capital cost) was 8.4 and 6.5 cents/kWh. A carbon tax would probably make nuclear more competitive, but it might not. In any event, it would help to be careful with numbers.

@Pavel - are you saying that my numbers were wrong? The numbers I was quoting were not projections, but actual numbers from current market and accounting data sources.

Yes, I was comparing the cost of fuel only, but when it comes down to it, both fission and combustion produce the same product - heat - which must then be converted into useful work using similar machinery.

The differences in costs between a nuclear heated steam plant, a coal heated steam plant, or a natural gas heated combined cycle steam plant are quite complex. There are features in each kind of plant that are unique and features in each kind of plant that do not need to be included.

In places where there is not carefully organized and well funded opposition, the capital cost differences are far less than they are in places where the opposition adds significant cost and risk to schedules.

The other factor that many ignore in discussing cost differences is the effect that the proven ability of nuclear power plants to operate at very high utilization rates (averaging in excess of 90% for the US fleet over a sustained period of a decade) has on the capital cost computation per unit of annual output.

I am quite careful with numbers and quite willing to engage in detailed discussions. None of my colleagues ever accused me of shyness.

Rod Adams
Publisher, Atomic Insights

Rod: Yes, I do think that comparing the cost of fuel only is a bit misleading. Of course, the cost of capital is a complex issue, but I'm not sure it depends that much on an organized opposition to nuclear power. For example, as I understand, opposition to nuclear power in Finland is not nearly as strong as in, say, Germany. But this doesn't help contain the cost of the new reactor construction. And you are not suggesting easing regulatory requirements to reduce the cost of new reactors, are you? I agree, predictability of the regulatory process would probably help reduce cost, but it does tell you something about the upfront capital requirements if regulatory delays could affect the overall cost that much.

I'm not an expert on the cost of the fuel cycle, but my impression was that the MIT study did take into account the high capacity factors demonstrated by the U.S. reactors. They tried to come up with the best-case scenario for nuclear, but still got it more expensive than coal or natural gas. Which, in itself, is not necessarily a problem - one could still imagine that in some circumstances nuclear would be a better choice. But if we ever make a choice in favor of nuclear power, we should do so knowing full well that it is more expensive and that there are some serious international security challenges that are associated with that choice. Advocates of nuclear power sometimes don't acknowledge these problems, which in the end, I believe, hurts their argument more than it helps.

@Pavel - The MIT study included a detailed appendix on their economic model and assumptions. There is no need to guess to recognize that they assumed a 40 year life, an effective interest rate that was nearly 2% higher than the interest rate assigned to natural gas plants, and that they assumed an average capacity factor of 85% for both nuclear plants and natural gas plants.

Those assumptions were certainly not favorable for nuclear energy and were not even terribly representative of reality. For example, even 30-40 year old nuclear plants routinely achieve CFs of 90% while the average CF for a gas fired unit in the US is well under 60%, even if you toss the peakers out of the sample.

I am not arguing that nuclear plants are "capital intensive" in that most of their costs are incurred during the initial building process. However, you apparently have no real understanding of the economic importance of imposed project delays in terms of adding the cost of carrying more people for a longer period of time, paying interest on non performing assets for a longer period of time, and delaying the time at which revenues can begin so that loan payments can be made.

I acknowledge that there are times and places where coal or gas plants can produce cheaper power than nuclear plants as long as you do not charge the fossil fuel plants for the service of allowing them to use our common atmosphere for a waste dump.

However, there are also many places in the US and in the world where there is no available gas pipeline and where transporting coal from the mine can double or triple the cost compared to the initial cost of the coal.

Those complexities are often ignored as well in economic discussions.

Rod: It may well be that under different assumptions than those made by the MIT study the cost of nuclear power would be different. You may be right that their scenario is not quite favorable to nuclear. But we are not talking about more favorable (or you would say, more realistic) assumptions resulting in nuclear being an order of magnitude cheaper than, say, coal (57 cents vs. $4.00 in the example that you used). My point was exactly that - if we discuss cost, then we should remember that in nuclear it is the capital cost that dominates very much everything.

On the cost of delays, I'm perfectly fine with admitting that I don't know the realities of nuclear power plant construction. I would just note, however, that if we are talking about delays at the licensing stage then I don't see how do you get expensive non-performing assets at that point. Unless, of course, you start construction before getting all the approvals. If the delay is about the quality of construction, then it's not quite the regulator's fault. Complexity of construction is also part of the (high) capital cost of nuclear reactors.

I don't understand all the fuss about demonstrating MOX fuel in commercial reactors. Many, many years ago (1960 and 1970) while working at NUMEC and Westinghouse I was involved in the successful production of MOX fuels for a several LWRs. If memory serves me they included the Saxton unit in Pennsylvania,San Onofer reactor in Southern California and a unit in Italy.

Yes, it would seem to be a proven technology. But for some reason Areva had problems with its fuel.

Leave a comment