Energy From Thorium Discussion Forum

I was wondering, with regards to the safety of LFTRs/MSRs. When the fuel goes down the drain plug, and gets channeled into the device designed for that purpose (what's it called?), so that it can cool, how much decay heat is there? With conventional LWRs, my understanding is that there is a great deal of decay heat for awhile (which lead to the problems at TMI and Fukushima).

I know that LFTR would be a little bit different (for example, it burns up the longer lived types of waste which are not burned up in LWRs), so it got me to wondering if the isotopic mix in the partially-used fuel would give off more, less, or about the same amount of decay heat?

All total slightly less but basically the same - roughly 5-7% of the full power level right after shutdown. The decay heat reduces fast during the first 24 hours. It continues to decay with time but the heat generation does not go to zero for a very long time.

The key here is that the reactor should have passive cooling. That is, the heat should be dumped to the atmosphere at a pace fast enough to keep the fuel cool. Normally, this is done by having sufficient passive heat removal to remove 0.5 to 1.0% of full power heat. For the first hours while the decay heat generated exceeds the passive cooling the fuel will heat up. The key here is to have sufficient heat capacity that the temperature rise is acceptable.

In a LFTR we pull out a significant portion of the decay heat when we continuously remove fission products to keep the neutronics good. Specifically, something like 1/4 of the heat comes out with the offgas system this heat doesn't disappear - rather we put it into a different location. We need to cool it there too.


On advantage we have is that we do not need deal with the neutron flux so that makes it easier.

My understanding is that the decay heat in a LFTR will be significantly lower (less than half) after 24hrs or so because most of the fission products are removed from the salt on a periodic basis. In a LWR the fission products remain in the fuel rods. Thus LWR have a much higher inventory of fission products and they are highly concentrated in a small physical volume.

The total decay heat will be the same. We will have some of it in the fuel salt, some in the off gases, some in the noble metals, and some in the extracted salt seekers. About half in the fuel salt after 24 hours sounds plausible, but we still have the other half - just in different locations.

Decay heat is energy, about 6% of the energy of nuclear fuel.

I would caution against removing too much of it, since most likely it converts energy into so called "waste."

More than 99+% of the time, this captured energy is used to our advantage. The public reaction to the events at Fukushima aside - which are, frankly, absurd reactions - I think the best thing to do with a reactor's fission products is to keep them in the reactor until they shut it down because of neutron poisoning.

As far as I know no other large reactor design can be passively cooled for more than a few days. The AP-1000 can go 72 hours before the tank on top of the containment building must be refilled. There are several squib actuated valves that must be activated although just once. LFTR on the other hand only needs passive heat pipes similar to the ones used on the Alaskan Pipe Line. When the core drains into the dump tank below the reactor on a SCRAM natural convection will have ~ 500ºc of thermal drive to cool the dump tank. No pumps, valves, squibs, or controls, just physics. All other sources of decay heat such as sparged gasses, and separated fission products can have similar heat pipe systems.
I believe that in the future government bodies will require a 30 day station blackout be a design case.

_________________
Mike Swift
Although environmentalists say we must reduce CO2 to prevent global warming they can never mention the “N” word as part of the solution.

I also think that we should not “waste” decay heat, however I think it would be a waste to allow fission products to shut down the reactor. If during salt processing we removed fission products, and placed them into small stainless steel cans we could store the cans, for maybe ten years, in a tank of coolant salt. This tank could be cooled by circulating coolant salt from the main loop, removing the decay heat, and placing it in the process. For a fail safe a number of heat pipes with bimetallic actuated valves could cool the fission product tank if the process load was lost.

_________________
Mike Swift
Although environmentalists say we must reduce CO2 to prevent global warming they can never mention the “N” word as part of the solution.

A 1 gigawatt nuclear plant produces about 1 ton of fission products per year, mixed with about 22 tons of uranium. That uranium is in the form of an oxide, so there is 3 tons of oxygen in there too, and some zirconium and traces of other things. The afterheat from that 1 ton of fission products decays as follows.



Some folks like the idea of burying the reactor underground, and allowing the afterheat to dissipate away into the surrounding rock. The trouble with this idea is that the ground is not a very good conductor. A 60 meter diameter hemisphere sunk into bedrock, with a surface 200 C hotter than the rock around it, can dissipate 75 kilowatts. It takes about 125 years for the fission products from one year's operations to decay this much. That's not much help in an emergency.

So the heat must be dissipated into the environment with a fluid, either water or air. Our 60 meter diameter hemisphere, this time poking up out of the ground, again 200 C hotter than the surroundings (air), will now dissipate around 8 megawatts. It takes about 9 hours after shutdown for the fission products to get down to that power level. That sounds like a more reasonable time frame.

In that time, the fission products have emitted 950 gigajoules of heat, which is a lot. That's enough heat to raise the hemisphere, if made of 40 cm thick concrete, to 200 C. However, containments are usually thicker than this, so it'll take a bit longer for the entire concrete dome to get up to sufficient temperature to convect away the afterheat, and the temperature it rises to will be a bit lower.

Lars, I used your decay numbers for this analysis, and I also checked against Kirk's ORIGEN results in his Java applet. Your numbers don't perfectly match, but it seems about right. If the integrated decay heat is right, why does AP1000 bother with a pool at the top of the containment? Air cooling would appear to work, so long as the heat can make it through the concrete.

The concrete itself is a bit of a challenge. The 60 meter diameter dome, made of 1 meter thick concrete, will transmit just 1.9 megawatts with a further 200 C drop across the concrete. This puts the interior surface at 400 C, which seems unreasonable. I'd want to embed heat pipes in the concrete if I were going to use it as my backup heat exchanger.

-Iain

Well, the chief nuclides that will do this in thermal spectra reactors are lanthanides, samarium, and to a lesser extent, europium.

Europium isotopes when concentrated may generate significant heat, but not enough to get anyone particularly excited since the nuclide is not a high yield.

I have looked into some detail means of direct extraction of lanthanides, and have convinced myself it can be done, but frankly, not much decay heat is going to be involved.

Most of the decay heat during operation is relatively short lived isotopes. The presence of I-131, for instance, shows that at Fukashima, <em>fresh fuels</em> failed. The good news is that this isotope is short lived. Within two months all of it will be gone. Other isotopes like say, Ba-140, and Ce-144, offer significant heat, but they are not real great neutron poisons.

Cesium is quite another matter. It can under exactly the right conditions volatilize. Because of many resonance absorptions, lighter istopes of cesium can be a drag on neutron economy during thermalization, but its not a big deal. There may be an argument for removing it from fluid phase reactors, although I'm quite fond of doing quite the opposite.

Most of the decay heat is left in the reactor since most of it is very short lived.
As to wasting some heat - this isn't terribly important. Heat itself is not expensive for us. The costs come with the pumps, heat exchanger, turbines, etc. If we waste 0.5% of the heat it does not change the amount we need to spend on the pumps, heat exchangers, turbines, or cold sink.

Let’s talk about decay heat in some detail and compare the oxide fuel cycle with a pure unity breeder LFTR. This is from memory and it has been a few years since I’ve run the numbers. Please correct me if I’m wrong.

Oxide fuel is kept in the light water reactor for about three years. 95 percent of the oxide fuel is U238. I’ve forgotten how much of the 238 is activated, but it contributes a considerable amount of energy to the reactor. Some of the energy is released through fission of the isotopes transmuted from U238. A great deal of energy is released through nuclear decay of the activated transuranics, and this decay heat is long lived. After a core dump, a LFTR running in a pure unity breeder mode should see very little heating from transuranic radioactive decay. LFTRs simply don’t make those isotopes because their fuel is not saturated with non-fissile heavy elements.

To be accurate, I’m ignoring the delayed neutrons released from fission. They are from radioactive decay, but we are talking about the sources of heat during the shutdown condition and the delayed neutrons are mostly gone after the first few seconds.

In a LFTR we strip out the gaseous fission products, so their decay heat is not in the reactor. Yes, we have to cool these fission products, but the important point is that their decay energy does not heat the fuel salt after shutdown and dump.

In most LFTR designs, some of the spent fission products are removed from the fuel. Sophisticated fuel cycles balance neutron economy and the life of the fission products. You would remove all the fission products to maximize the amount of 233 produced. In contrast, if you wanted to quickly reduce the radioactive inventory, then you would leave the long lived fission products in the fuel to transmute them into short lived isotopes. This means a LFTR might not have many fission products in the fuel at any one time, depending on the processing rate of fission product removal. We might have almost no decay heat, or we might have the full compliment of fission products to cool after shutdown depending on the fuel cycle.

In contrast to conventional oxide fuels, we don’t have to cool the transuranics, the gaseous fission products, or much of the fission products in the fuel dump tank. That helps a lot. It also keeps the cost burden down because we can spend more capital on making power rather than emergency equipment. We have backup systems, but ours are potentially smaller and simpler than other designs.

_________________
It is good to be splitting atoms again on the weekend.

The vast majority of the heat (93%) is from the fission itself. This dies off rather rapidly once the reactor is shutdown.

Almost all the rest of the heat is from decay of the fission products. Almost none of that is changed by neutron absorbtion. Almost none of it is from alpha decay of actinides.

In removing some of the fission products we do reduce the heat load in the reactor. The heat is moved to where ever we store the fission products.

The advantage we have over LWRs is primarily that there is no motive force to drive our fission products into the atmosphere. We have roughly the same decay heat load to deal with. Since we have moved some of the heat load away from the fuel and can design our cooling system without the constraints of being transparent to neutrons and surviving a neutron flux our job is easier. I fully expect any LFTR will have a fully passive means to deal with decay heat.

I claim that dealing with the decay heat load is something that has been progressively better addressed in more recent designs.

The delayed heat load from alpha decay of actinides can be dissipated in a gradual and ongoing on-the –fly cooling process to the ambient air using updraft airflow provided by high chimneys away from the core salt. Most heat from alpha decay of actinides will be released in a month or less leaving little decay heat during any dump tank cooling.

_________________
The old Zenith slogan: The quality goes in before the name goes on.

Actinides are not your near term heat source.

I quite like heat pipes as a simple pump free heat transport mechanism. And while sodium is no doubt is an excellent fluid for a decay heat system heat pipe, but if the ultimate heat sink is boiling water, that would be a bad combination (water/sodium).

I note that NaF-NaBF4 has a melting point of 385C and boiling point of 694C (ORNL TM-2009-69 Table 2). I also note that at 900C it has a vapour pressure of 9500 mmHg (12.7 bar).

Given the combination of Hastelloy compatibility, mp, bp and vapour pressure, can we use NaF-NaBF4 as a heat pipe fluid in a passive cooling system?

It sounds ideal from a thermophysical property point of view, having boiling point in a range of 694 - 900C and low viscosity seems promising for high temperature heat pipe application.