Saturday, April 18, 2009

The French View on Graphite in MSRs

The potential scalability of the LFTR is a powerful argument for its adoption as an energy solution or the energy solution. There are two primary limits of LFTR scalability. They would be
1. The ability to produce LFTR's with at leasr a one to one conversion ratio. That is the bility to produce as much nuclear fuel as they consume
2. The availability of as much fissionable start up charges as would be required by large scale deployment.
The conversion ratio would be a matter of LFTR design. But there is a potential constraint on the required input of fissionable materials to start the LFTR. In all reactors moderators act to increase reactivity. The best moderators are heavy water and graphite. Early reactors which used natural uranium as fuel required either graphite or heavy water as moderators. Light water is a significantly less powerful moderator than either graphite or heavy water. Graphite moderated reactors require a significantly smaller start up charge - perhaps 25% or even less fissionable material to achieve criticlity. Unmoderated reactors require far more fissionable material in order to maintain a chain reaction. It LFTRs the carrier salts function as partial moderators. Thus it is possible to operate a LFTR without graphite, but its opperation will require a far larger inventory of fissionable materials. That means more fissionable materials for the start up charge. Thus the use of graphite would be an emportant componant of LFTR scalability. It would be possible to build a large number of LFTR's without but that would require consideably more enriched U-235 or more U-233 or more Pu-239. Thus the use of Graphite in LFTR's would be highly desirable.

Recent French research, however, has pointed to problems or at least questions about the use of graphite as a moderator or structural material in MSRs. ORNL researchers were aware of the effects of neutron radiation on graphite. As The French researchers noted:
This concerns in particular how the graphite reacts to irradiation exposure. Beyond a certain degree of damage, it becomes the seat of swelling. Graphite’s life span is determined bythe time it takes to reach a fluence limit,
In addition to the well known problem with graphite swelling, French researchers (L. Mathieu,D. Heuer, R. Brissot, C.LeBrun, E. Liatard, J.-M. Loiseaux, O. Méplan, E. Merle-Lucotte, A. Nuttinand, J. Wilson, C. Garzenne, D.Lecarpentier, and E.Walle) believe that they had spotted another problem with graphite. They descrabed the classic ORNL graphite moderated MSBR:
Our standard system is a 1 GWe graphite moderated reactor. Its operating temperature is 630 C and its thermodynamic efficiency is 40 %. The graphite matrix comprises a lattice of hexagonal elements with 15 cm sides. The total diameter of the matrix is 3.20 m. Its height is also 3.20 m. The density of this nuclear grade graphite is set to 1.86. The salt runs through the middle of each of the elements, in a channel whose radius is 8.5 cm. One third of the 20 m3 of fuel salt circulates in external circuits and, as a consequence, outside of the neutron flux. A thorium and graphite radial blanket surrounds the core so as to improve the system’s regeneration capability. The properties of the blanket are such that it stops approximately 80 % of the neutrons, thus protecting external structures from irradiation while improving regeneration. We assume that the 233 U produced in the blanket is extracted within a 6 month period.

They note the importance of the size of channels in the graphite:
The size of the channels in which the salt circulates is a fundamental parameter of the reactor. Since the size of the hexagons is kept constant in all of our studies, the size of the channels determines the moderation ratio. Changing the radius of the channels modifies the behavior of the core, placing it anywhere between a very thermalized neutron spectrum and a relatively fast spectrum. The cross section resonances of the materials present in the corehave a strong impact on the neutronic behavior of the reactor.
They noted something that appeared to have escaped ORNL researchers. As the heat of graphite increases a positive coefficiency or reactivity effect,
comes from an energy shift of the thermal part of theneutron spectrum(around 0.2eV), due to heating of the moderator. This shift increases the fission rate because of a smalllow energy(0.3eV) resonance in the fission cross section of 233U. It simpact on the stability decreases as the amount of graphite in the core decreases and as the influence of th ethermal portion of the spectrum weakens.
Now if you are a really cool cat, you know that this means that there is a potential reactor safety problem related the heating of graphite in a reactor in which U-233 is being burned. Sacre bleu! But getting rid of the graphite helps. But what the French researchers are really trying to say is, if you build a 1 GW Graphite moderated MSR and run it at full blast, you are going to shorten the lifespan of the graphite moderator, by subjecting it to a lot of neutron radiation and heat. Furthermore heating the graphite creats an effect that makes the reactor less safe.

The French researchers tell us:
Since the safety aspect cannot be circumvented in the design process of a nuclear reactor, we consider that this constraint is necessarily satisfied. Moreover, we consider only those configurations whose total feedback coefficient, not just the salt feedback coefficient, is negative. Except in the case where the size of the reactor is reduced dramatically, leading to a significantly increased neutron flux, the total feedback coefficient is negative only for either very thermalized or fast neutron spectra. The first option implies a small fissile matter inventory and a weak neutron flux. When submitted to such a flux, the graphite undergoes little damage and its life span is reasonably long.

On the other hand, captures in the moderator deteriorate the breeding ratio significantly. If a reactor system does not need to regenerate its fuel, then this very thermalized configuration may be suitable.
Thus the French researchers tell us:
Decreasing the specific power of the reactor (by increasing its size and/or decreasing the total power generated) leads directly to a decreased flux intensity and, as a consequence, extends the graphite’s life span
They add:
This, however, increases in the same proportion the per GWe fissile matter inventory, without providing a very satisfactory solution.
Now ir sounds like the French want to get rid of the graphite real bad. But note that their study focused on a 1 GWe reactor - the MSBR - opearating at 630 C. But there is an exception if the reactor were dramatically smaller. I wonder if say 100 MWe would be considered dramatically smaller in France.

The French operate their electrical grid with a bunch of huge reactors. They do not run air conditioners in the summer, and if there is an August heat wave, a whole lot of old French men and women are going to die. In Texas we value our old people, and want to keep them around, so we have a big summer electrical reserve. I maintain that any acceptable solution to the energy crisis will keep my air conditioner running all through the Dallas summer. In Texas now the reserve power generators run on natural gas. it would b nice to switch that system to nuclear. Big French reactors won't do the trik, because they cost tooo much, but little LFTR's that generate 100 MWe would do the trick, and they don't have to run real hot for efficiency. It would be a blessing if Kirk Sorensen would design one that could be built and operated at a low price. Now Kirk has not spelled out details to me, but I'll bet he has got a few tricks up his sleave, that would give a graphite moderator a reasonably long life, and will take care of those safety issues that have so worried the French.

If you want to understand the writings of French scientists, you should read Bruno Latour. Latour notes that political elements are never very deeply burried in the text of French scientific papers. One always needs to look for wiggle room in reports of scientific research. The French have left wiggle room in their statements related to the use of graphite in small LFTR type reactors. Considering the advantages I for one am not ready to put paid to the graphite moderated Big Lots reactor yet.

1 comment:

donb said...

David LaBlanc comments on a simple fix to the positive coefficient of reactivity problem here at the "Energy from Thorium" web site.


Blog Archive

Some neat videos

Nuclear Advocacy Webring
Ring Owner: Nuclear is Our Future Site: Nuclear is Our Future
Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet
Get Your Free Web Ring
Dr. Joe Bonometti speaking on thorium/LFTR technology at Georgia Tech David LeBlanc on LFTR/MSR technology Robert Hargraves on AIM High