Ask Atomic: What limits reactor fuel burnup?
Several people have either sent email or made a comment on the Atomic Insights Blog about the factors that limit reactor fuel consumption (burnup). Here is one of the questions with my answer.
This question came from Iain, who blogs at http://ambivalentengineer.blogspot.com
Rod, Could you please explain something to me, or point me at some document that explains it?
One way to reduce the amount of used fuel is to get higher burnup of the fuel used. I’m trying to understand what limits the burnup of fuel in commercial reactors. If ordinary reactors burnup 6% of the fissile and fertile material, you’d think that a reactor that burned 50% of the fissile and fertile load would be able to run for decades without refueling (the last decade just on the decay of fission products alone).
So far, I understand that as the fissile material fissions, it releases neutrons, some of which create more fissile material, and fission products, some of which are poisons which tend to absorb neutrons without further fissions. I think I understand that for a given spectrum of neutron energies, the reactor core has either negative, neutral, or positive reactivity, which means the average absorbed neutron leads to either fewer than one, one, or more than one neutron, after accounting for all the various losses. As the fuel is burned, reactivity eventually drops as the poisons accumulate and absorb a significant fraction of the neutrons.
So at first blush it looks like the core burns until reactivity drops below neutral. But the core is arranged so that a positive reactivity portion (controlled to be neutral) is surrounded by lots of other stuff, which can absorb excess neutrons. I assume that operators tend to put fuel burned to negative reactivity in this periphery, so that the neutron flux can cause further reactions and burn even more of the fissile material, causing the reactivity to drop even further.
I know that CANDU reactors use heavy water to multiply the neutron flux so that they can operate with lower reactivity, and thus unenriched natural uranium. Would the use of heavy water allow a reactor burning 4.4% enriched uranium to achieve higher burnup? How do I calculate how much more burnup you could get? (I don’t actually expect an answer for this last question 🙂
So what is the burnup limit, and do operators currently burn fuel all the way to the limit?
You have covered pretty well the factors that influence reactivity. As you may or may not know, there are reactors in operation today that run for decades without new fuel because of careful attention to some of the factors that you mention. The Virginia class submarine, for example has a core that will last for the life of the ship – approximately 30 years. There are many other longer lived designs, often incorporating one version or another of the seed and blanket concept used in the Light Water Breeder Reactor program.
One minor correction to your initial comment above – no matter how long you run a reactor, you will not be able to obtain a large portion of the heat from fission product decay. Roughly 6% of the heat produced by a reactor is from decay of fission products, but the vast majority of that is from very short lived isotopes that disappear rather quickly after fission stops. An hour after shutdown, decay heat is approximately 1% of the power produced by fission during operation and it continues to drop as the decay continues.
In addition to designing the core to keep reactivity high enough to sustain a chain reaction, fuel designers and operators have material limits on how much fuel they can ultimately burn. A basic requirement of reactor design is to keep fission products and heavy metals contained in the fuel and out of the coolant.
Fuel elements are in a challenging environment with large heat production, an intense neutron and gamma radiation flux, and exposure to coolant flow. The fission process releases some gases, both as fission products and because most of the fuels used are oxides that have a couple of oxygen atoms in each molecule. When the heavy metal fissions, some of those oxygen atoms are set free. This gas production creates internal pressure in the fuel element. All of these stresses on fuel elements have an effect on the ability of the element to maintain its integrity and its mechanical properties.
In order to maintain the primary imperative to keep the fission products and the fuel materials from entering the coolant, the ultimate burnup of the fuel is kept below failure limits that have been determined by an extensive testing program in specially designed and operated reactors. It takes a long time to run these burnup test programs – it is difficult to accelerate them more than 2 or possibly 3 times as fast as they would occur in a normal reactor.
Since current commercial reactors replace 1/3 of their fuel every 18 months to two years, fuel usually stays in operating reactors for up to six years. As you can imagine, the timelines to test improvements start to stretch a long way, especially since the test reactor availability is quite limited. There is a definite cost for such programs, and there is not very much payback for the fuel suppliers if they undertake efforts to increase the longevity of their fuel. In fact, in an environment where the number of reactors is constant, there is an economic disadvantage because longer lived fuel means a reduced sales volume. Anyone can understand why fuel suppliers have decided that their current burnups are good enough.
The destruction of the Fast Flux Test Facility (FFTF) by the Department of Energy did not help to encourage any new fuel testing since the lines at the rest of the available test reactor(s) is now longer. That action, first approved during the Clinton Administration, but enacted by the Bush Administration, will soon become recognized as a huge mistake.
Possible path for improvement
With TRISO particle based fuel – either pebble bed, prismatic, or some other fuel form that has not yet been suggested – there is a body of evidence that points to an ability to reach very high burnups and extended core lifetimes. Part of the advantage is the fuel form of the particle, part of it is in the fact that the fuel does not come in a form where the fuel itself is part of the structure. In a traditional light water reactor, the fuel is encased in skinny tubes with thin zirconium cladding that are 12-15 feet long. The stacks of heavy metal oxide pellets that fill the inside of each of those tubes is part of the structure that keeps the tubes from deforming.
I hope this helps understand the scope of the challenge and even to recognize some of the paths for advanced fuel cycles that will alleviate some of the issues that limit reactor fuel burnup in today’s reactors.