Monday, August 25, 2008

Reactivity-Initiated-Accident Analysis without Scram of a Molten Salt Reactor

Journal of Nuclear Science and Technology
Vol. 45 (2008) , No. 6 p.575-581

Reactivity-Initiated-Accident Analysis without Scram of a Molten Salt Reactor
Nobuhide SUZUKI1) and Yoichiro SHIMAZU2)
1) Mitsubishi Heavy Industries, Ltd
2) Graduate School of Engineering, Hokkaido University

Recently, a conceptual design of a small MSR, named FUJI-12 has been proposed. FUJI-12 operates with the same fuel salt as the Molten Salt Breeder Reactor (MSBR) designed by ORNL. The authors are interested in the MSR concept due to its high potential in the areas of safety, proliferation resistance, resource sustainability and waste reduction, all necessary requirements for the generation IV nuclear power systems. The authors believe that additional investigations are necessary for future study. From this point of view, the authors have analyzed various reactivity insertion accidents due to control rod malfunctions in FUJI-12. The MSR can be operated with a small excess reactivity. However, at the same time, the delayed neutron fraction is quite small due to the usage of U-233 as fissile material and the circulation of the fuel salt. Therefore, the reactivity insertion accident should be qualitatively evaluated. The reactor transients were analyzed without scram in order to evaluate the severity of such accidents against the safety. Although the total primary system design of FUJI-12 is not completed, and thus, the accident analyses include some crude assumptions, it can be expected that the reactivity insertion accident in FUJI-12 would not result in severe plant conditions.

We have analyzed reactivity-initiated accidents at hot
zero power and full power conditions without scram in the
Molten Salt Reactor, FUJI-12. The maximum temperatures
for inlet and outlet fuels were same at zero power condition.
They were 950 and 1050 K, for one and two control rod
withdrawals, respectively. When three control rods were
withdrawn, the maximum temperature was 1180 K. The
safety limiting temperature for the inlet fuel is 1050 K, and
for the outlet fuel, 1200 K. Thus, it can be concluded that
the safety limit is not violated for up to two control rod
insertions at zero power.
The maximum fuel temperature for two rod insertions of
0.344%dk/k was 1220 K at the rated power. This result vio-
lates the safety limit. However, as discussed in the previous
section, the maximum reactivity required for the normal
operation is expected to be less than 0.2%dk/k. Thus, the
reactivity addition of two control rods of 0.344%dk/k is
hardly expected. Therefore, it is concluded that the reactivity
addition accident during normal operation will not result
in safety violation. In other words, in order to ensure safety
for reactivity-initiated accidents, the reactivity addition must
be limited to less than that of two control rods worth. For
example, the utilization of smaller sized control rods should
be considered. If it could not be adopted, it could be said
that some mitigation system might be installed to protect
against the simultaneous control rod insertion. The other
system to protect the reactor is an automated or passive salt
drain system that has been designed for the MSBR.


Anonymous said...

Ideally, in a MSR design, the heat exchangers would be inside the reactor vessel. In that case, all the fuel inventory stays in the reactor and the delayed neutron fraction can be fully used to control the reactor. Control bars would be well nigh superfluous outside of startups and full shutdown conditions. The reactor would be self-controlled, purely by temperature.

Considering the temperatures molten salts can reach safely (1,600K or so), the temperature difference between the primary (fuel salts) and the secondary could be quite high, hundreds of K if needed. Then, with a big temperature delta, the heat exchange surface can be pretty small, may be even reduced to the outside walls of the reactor itself with some ribbing to get a larger exchange surface and relief stress.

The drawbacks would two folds. 1) The heat exchanger would have to withstand a large temperature differential on its walls, which is mechanically very stressful. Probably a very hard nut to crack 2) It would expose the 'secondary' fluid to high neutron flux and activate it. But, contrary to 1), I'm not convinced it's such a big deal. Even with the exchangers outside, secondary activation is still there, 2 or 3 orders of magnitude less but there nonetheless (from the delayed neutron, as a matter of fact). So it has to be managed anyway.

Even more provocative, I remember reading about direct contact heat removal for ultra-compact fast neutron MSRs, essentially a shower of molten lead in the core across the salt: pour from the top, collect at the bottom with gravity or centrifuge separation between lead and salt. Then circulate the lead in the exchanger properly said. Wicked stuff!

Of course, behind all of that, the real problem for MSRs remains neutron fluence at high temperature and chemical reactivity. Metals can't cut it (plus the issue with RE and platinides plating). Even graphite is too susceptible to neutron damage. Carbon-carbon composites? Silicon carbides?

Anyway, one thing is certain. This technology really needs a big investment in material science. The pay-off would be huge.

Charles Barton said...

I think ORNL looked at the use of Molten metal fluids as coolants flowing through the fluid salts, but not mixing with them. ORNL did not commit to the idea. Heat exchanges could be nested between the primary reactor wall and the blanket in a two fluid MSR design. ORNL materials scientists believed that they had beaten all of the materials problems, including the problem except for graphite before its research was shut down in the mid 1970's. They believed that they had the problems with Hastelloy under control. There are a number of tricks that can mitigate the graphite problem, and the French have decided to research a graphite free design. Carbon-carbons might work if the same tricks that work for graphite also work for them. Silicon carbides might work but need research.

Finally the problem with tritium can be solved. One approach would be to use an alternative fluoride to LiF in the coolant/fuel formula.

Kirk Sorensen said...

Hey Charles, there's other ways to beat tritium than replacing LiF in the salt. LiF is too good to give up on--low neutron absorption, rock solid chemical stability, good low-melting point combinations.

The place to catch tritium, in my opinion, is in the gas loop of the gas turbines. If there's any oxygen introduced to the helium, CO2, or nitrogen used as working fluids, the tritium will react with the oxygen forming tritiated water which can be recovered on sorbent beds.


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