Energy From Thorium Discussion Forum

It is currently Oct 16, 2018 5:00 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 192 posts ]  Go to page Previous  1 ... 9, 10, 11, 12, 13
Author Message
PostPosted: Jul 09, 2010 6:00 am 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2246
A suspension, even a nano-particle based one, presents its own problems. Even if we find advantage in a fluid fuel, the blanket can be replaced by a 3-dimensional grill of thin sheets or wires. Phase separation would be enough separation. There would be very little fission in the thorium 3-D grill which will mainly absorb the neutrons. The fission will be mainly confined to the liquid, with easy escape of volatile fission products. A reflector-cum-blanket of thorium metal sheet cladding on the reactor walls would also afford protection from fluoride salt solution by sacrificing part of the blanket, whatever goes into solution.
The solid thorium can be replaced every few years and electro refined to recover U-233. Thorium can be recycled.


Top
 Profile  
 
PostPosted: Jul 09, 2010 11:25 am 
Offline

Joined: Dec 01, 2006 7:23 pm
Posts: 141
Lars wrote:
We generate 2400 curies of the stuff per day. Sounds like a lot - especially if I express it as 2.4 billion, million, pico-curies like the newspapers might report it. In reality it is 246 milli-grams. The legal emissions limit is 10 curies but the Vermont-Yankee furor was over 1 milli-curie/day release.


This is a ridiculously small amount. Here is a label from a tritium "Exit" sign: http://www.orau.org/ptp/collection/radioluminescent/h3exitsignl.htm , it has 25 curies. Wikipedia claims industrial consumption of tritium as 400 grams/year (3.9 million curies). All I can say is they better never break an "Exit" sign.

_________________
Paul Studier


Top
 Profile  
 
PostPosted: Jul 09, 2010 11:52 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3063
And yet they managed to derail (hopefully just temporarily) renewal of the operating license for Vermont-Yankee over this issue. I'm thinking we will have to meet the as low as reasonable attainable standards. Hence, the desire to remove as much tritium in the off-gas as possible. It seems likely that we can remove T from 100% of the extracted off-gas (rather than the 20% planned by ORNL) with very minor expense provided we find a way to separate Kr+Xe from T+He. Since there is a very large mass difference I'm thinking it should not be hard to separate them.


Top
 Profile  
 
PostPosted: Jul 09, 2010 2:53 pm 
Offline

Joined: Dec 01, 2006 7:23 pm
Posts: 141
Yes, it is tragic to shut down a nuke which will keep a coal plant operating. It seems to me that T could be oxidized to water and soaked up with a desiccant such as silicon gel. Krypton boils at 119.93 K, Zenon at 165.03 K, so a liquid nitrogen cold trap at 77.36 K should trap those easily, along with any tritium not captured by the desiccant. Then for storage, one would have to store them as compressed gas.

_________________
Paul Studier


Top
 Profile  
 
PostPosted: Jul 09, 2010 3:43 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3063
Yes in the end we can use a liquid nitrogen trap to separate Xe from Kr from He for long term storage of the Kr. BUT what I'd like to do is to take the hot off-gasses and immediately separate the Xe+Kr from the T+He. At this point the Xe+Kr are throwing off lots (20 MWatts) of decay heat. I don't think we want to pump that through a liquid nitrogen trap. The purpose of the early separation is two fold.
1) it makes it practical to store the Xe+Kr as a gas until the heat production is low enough that we can send things through the liquid nitrogen trap to finish separating He, Xe, and Kr.
2) it makes it practical to send the T+He through the CuO tubes to grab the tritium as T2O.

The early separation does not need to be anywhere near perfect. It starts at 0.1% Xe+Kr. If after separation we have 10% Xe+Kr and 90% He then we have reduced our gas storage volume 100x. If we can keep 99% of the Xe+Kr out of the T+He stream then we have reduced the decay heat 99% in the T+He stream.

Given the differences in mass T2 at 6 and Kr at >80 it seems like we should be able to separate these two mechanical pretty easily.


Top
 Profile  
 
PostPosted: Jul 09, 2010 5:29 pm 
Offline

Joined: Jul 28, 2008 5:01 am
Posts: 461
Location: Teesside, UK
Lars wrote:
........The early separation does not need to be anywhere near perfect.......Given the differences in mass T2 at 6 and Kr at >80 it seems like we should be able to separate these two mechanical pretty easily.
For such quick-and-dirty separations, the usual first choice is pressure-swing adsorption. We know from ORNL work that Xe and Kr will stick to carbon, so there is a good chance it will work. Energy requirements are small compared to cryogenic systems - indeed, since the 20 Mw decay heat will get dumped into the gas stream at the high-pressure end of the cycle, you could decompress through a turbine and couple that to the compressor, and get it to power itself.


Top
 Profile  
 
PostPosted: Jul 10, 2010 12:24 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3063
As the Xe decays it will deposit material into the adsorber causing it to lose capacity, eventually become a radioactive waste stream, and posing a decay heat challenge.

The ORNL approach uses a 2 hour holdup into the dump tank. This lets most of the decay happen there and the decay products simply collect in the bottom of the tank (and end up getting mixed back into the fuel salt - something I think we should avoid). Even still, the carbon adsorber gets clogged over time and becomes a significant radioactive waste stream.

I am hoping that some technique can exploit the mass difference in a process that won't generate additional waste. Something like a pitot tube or sending the gas at high velocity through a coiled pipe so that the heavy gas (and any decay liquids) gravitates to the outside.


Top
 Profile  
 
PostPosted: Jul 13, 2010 4:08 pm 
Offline
User avatar

Joined: Aug 21, 2008 12:57 pm
Posts: 1056
Quote:
    Overview of dimensionless numbers in fluid dynamics

    • Nusselt number: Ratio between total heat transfer in a convection dominated system and the estimated conductive heat transfer. For Liquid Lithium, NUSSELT NUMBER = 3.65. For comparison with FilBe, NUSSELT NUMBER = 30 to 70 increasing with temperature.


    • Grashof number: Ratio between buoyancy forces and viscous forces.


    • Prandtl number: Ratio between momentum diffusivity and thermal diffusivity. Typical values are Pr = 0.01 for liquid metals; Pr = 0.7 for most gases; Pr = 6 for water at room temperature. The Prandtl number for lithium is 0.05 and 0.004 for sodium at 649C. For comparison with FilBe, the Prandtl number is 13.525.


    • Rayleigh number: The Rayleigh number governs natural convection phenomena. It is the ratio between inertial and viscous forces. The product of the Grashof number and the Prandtl number gives the Rayleigh number, a dimensionless number that characterizes convection problems in heat transfer. When the Rayleigh number reaches a critical value then turbulence begins.

    • Péclet number: The Péclet number is a dimensionless number relevant in the study of transport phenomena in fluid flows. It is defined to be the ratio of the rate of advection of a physical quantity by the flow to the rate of diffusion of the same quantity driven by an appropriate gradient. In the context of the transport of heat, the Peclet number is equivalent to the product of the Reynolds number and the Prandtl number.


    • The Strouhal number: The Strouhal number is a dimensionless number describing oscillating flow mechanisms in dimensional analysis.




jaro wrote:
Sorry, but there seem to be way too many unknowns in this concept!


Agreed.

Let us look at this in some detail and try to get a handle on some of the unknowns. I think the key to riding this horse is to take fullest advantage of lithium as a coolant.

Among the liquid metals, lithium has superior overall thermo-physical and electrical properties for a reactor coolant. Lithium has the highest heat capacity, close to that of water, high thermal conductivity and low Prandtl number.

Being a liquid metal, lithium boasts high thermal conductivity from its metallic nature equal to silver, with large thermal diffusivities. Lithium is a good coolant and it is one of the reasons for the use of liquid metals in applications with high energy fluxes and high power densities.

These good coolant properties are enhanced by its high volumetric expansion coefficient, which lends itself to applications with high Grashoff number in which natural convection can play an important role.


Free convection is an appreciated feature from the point of view of safety considerations to ensure coolability.


But in this application, convection is not so important in how heat is distributed around the reactor vessel but more in how heat is removed from the immediate vicinity of the individual particles which transfer heat to the coolant is primarily through conduction and not convection.

The Prandtl number relates the dynamic viscosity of the fluid and its thermal conductivity. Lithium’s Low Prandtl number implies that turbulent effects have far less importance than heat conduction in the transfer of heat.

Low Prandtl number fluids are characterized by their difficulty in developing turbulence by its relative high dynamic viscosity and heat transfer is very much affected by thermal conduction, given a flow pattern in which temperature gradients are important. This is also a consequence of its low Peclet number.

In this concept, turbulence is a minor mechanism for the transference of heat throughout the reactor coolant. In order to reduce mixing of the seed and blanket particles as much as possible, a design objective is to suppress turbulence in the coolant to a major degree.

jaro wrote:
Per my earlier post, I believe that "separation via characterization of seed and blanket particles" would in fact be a large fraction of the total particle flow (ie. near 100%), since the rate of particle mixing that occurs between the core and the blanket zones would be very rapid for any sort of practical system with adequate heat transfer characteristics....

In a minced-meat viscous type of slurry, the heat can't get out and the particles would destroy themselves in no time!



The goal is to keep the seed and blanket completely separate and I believe this is possible because of the unique physical and coolant characteristics of the lithium coolant.

A minced-meat viscous type of slurry is not practical or even possible with lithium coolant. Lithium transfers heat mainly by conductance because it is a metal, and a superior one for heat transfer at that. The goal is to establish a rising laminar heat plume in the core were the seed particles are buoyant in lithium and rise in a straight line riding at zero relative velocity vis'-à-vis' the lithium coolant.

The core and the blanket are kept apart by an annular shaped separation zone in which seed and blanket particles do not enter. The width of this zone is proportional to the magnitude of the horizontal vibration component of the ascending particles.

Natural convection is totally suppressed by forced circulation pumping. This eliminates the possibility of the Rayleigh number from reaching a value that triggers turbulence. This insures that a laminar heat plume is maintained in all situations.

Another thing that could be done is to apply a week magnetic field in the Z or X direction to reduce the Strouhal number of the particles. For and explanation of this technique see slide 6 of the following:

http://www.fzd.de/workshops/sino-german ... resden.ppt

This slide illustrates how a longitudinal magnetic field can be applied to a particle to eliminate the turbulent wake of that particle so that it ascends in a straight line path through a liquid metal.


jaro wrote:
It seems to me that you can't have it both ways:
Either you have a nice turbulent slurry supension that mixes very rapidly (between core and blanket particles), making separation via characterization of seed and blanket particles very difficult due to the rapid turnover.
Or you have a minced-meat-style viscous fluid that maintains separation in the core, but is hard to pump through heat exchangers -- where mixing occurs anyway -- and which makes subsequent characterization of seed and blanket particles very difficult because of the fluid viscosity...


Lithium may enable a third alternative. By using the unsurpassed heat conduction character of the lithium coolant, a laminar vertically rising heat flow mechanism transports heat to the top the reactor vessel for removal thereby constraining advective particle ascent where mixing of particle types does not usually occur.

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


Top
 Profile  
 
PostPosted: Jul 13, 2010 7:10 pm 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1946
Location: Montreal
Excellent post Axil !!
Thanks for taking the time to review all those named parameters -- its been quite a few years since I've had to do calculations with those 8)
Your patience is really quite astounding sometimes !

Axil wrote:
The goal is to establish a rising laminar heat plume in the core were the seed particles are buoyant in lithium and rise in a straight line riding at zero relative velocity vis'-à-vis' the lithium coolant.

I like lithium coolant.
Many years ago, I chose it as the prefered option in a particular reactor application -- in an undergrad research report.
However, like other metals, liquid lithium -- especially at high temperature -- is a LOW viscosity liquid.
I'm not sure what exactly you need to do to get laminar flow with a slurry of liquid lithium and nanoparticles, but low flow velocity is probably a big part of it.
In fact, the bulk flow velocity of the slurry might need to be so low, that the temperature increase from the inlet of the reactor core, to the outlet, might be unaceptably high, given a reasonable volumetric power production rate (say 400 kW/litre ?).
Perhaps you need to plug in some numbers into that list of parameters, and see if what comes out makes any sense ?


Top
 Profile  
 
PostPosted: Jul 14, 2010 3:58 am 
Offline

Joined: Dec 26, 2007 11:45 am
Posts: 191
Lars wrote:
Yes in the end we can use a liquid nitrogen trap to separate Xe from Kr from He for long term storage of the Kr. BUT what I'd like to do is to take the hot off-gasses and immediately separate the Xe+Kr from the T+He. At this point the Xe+Kr are throwing off lots (20 MWatts) of decay heat. I don't think we want to pump that through a liquid nitrogen trap. The purpose of the early separation is two fold.
1) it makes it practical to store the Xe+Kr as a gas until the heat production is low enough that we can send things through the liquid nitrogen trap to finish separating He, Xe, and Kr.

What if gaseous fission products could be separated without helium?

Pump some of the salt up a relatively short column (a few meters) to a vacuum chamber. Put it in a shallow dish and agitate the surface with strong ultrasonic vibrations. This will cause micron-sized droplets to form above the surface of the dish. They won't get carried away as aerosol because the pressure is too low but this should generate a very large surface area for the gasses to get out of solution during the droplet's ballistic flight. The pumps scrubbing this chamber will produce a stream of almost pure Xe+Kr (+traces of T).

I wonder if this scheme can be made effective enough to compete with helium sparging without requiring excessive out-of-core salt volumes.


Top
 Profile  
 
PostPosted: Jul 18, 2010 2:34 pm 
Offline
User avatar

Joined: Aug 21, 2008 12:57 pm
Posts: 1056
I’m wondering? What is the fundamental reason(s) for establishing a seed and blanket configuration where the seed zone in a reactor vessel is physically separated from the blanket zone? Is it a matter of optimizing neutronic efficiency and reducing neutron losses; if so, what is the cost in neutron losses by using a 1 ½ fluid solution to the seed and blanket zone wall conundrum? If these losses exist, can these be mitigated for in some other area or technology?

Does it make seed and blanket material identification, isolation, and processing easier? Or is it just a tradition in that it has always been done that way?

What are the fundamental tradeoffs involved? Can anybody help?

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


Top
 Profile  
 
PostPosted: Jul 18, 2010 3:16 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3063
The seed section is where the fissile material is and where the reactor power is generated. It is here where the concentration of fissile is high enough that the reactor is critical. The neutrons travel some distance before being absorbed. Some of those being absorbed cause fission and so launch another set of neutrons. If the material is uniformly distributed then the density of neutrons is roughly cosine shaped with the highest density in the center of the reactor and decreasing to "near" zero at the edges. This is primarily a geometry problem - it has little to do with the neutron spectrum or fuel used or whether there is fertile in the fuel or not. We would love it if the neutron density at the edge of the reactor were actually zero - life would be much easier then. It isn't so we try different approaches.

One way is to simply let the reactor end at the edge - this results in the neutrons leaving the reactor, damaging the wall, and possibly generating a neutron flux outside the wall. Not acceptable.

Another way is to put "reflectors" around the reactor. While this sounds great (and is the normal practice) it means that we have a higher density of neutrons near the edge - which means more fission near the edge and more neutrons get generated there which means more neutrons hit the reflector etc. While reflectors are good to reduce the size of the reactor required to achieve criticality it isn't clear that they reduce neutronic losses. This is especially so when the reactor is faster and the reflector also is a moderator (like a graphite reflector).

A blanket is normally the outer section of the reactor with no fissile and lots of fertile. Neutrons leaving the core then either get reflected by the fertile (and its salt) back to the core or get absorbed by the fertile. A few neutron will penetrate the blanket as well and so we need to have an absorber at the edge of the blanket to reduce the neutron flux outside the reactor. Four times as many neutrons are captured in the blanket as leave the reactor in the French design (with a 0.4m blanket). Other designs increase the blanket thickness and thus achieve even lower neutron losses to leakage. Note that a blanket does definitely help improve the neutron economy.

In ORNL's single fluid design they had lots of graphite moderator in the core so that most of the fission happens as a result of slow neutrons. Along the edges they had less graphite so that most neutrons did not slow down as much. Given the same salt content a thermal neutron is much more likely to cause fission than a fast one. This is because the fissile atoms fission cross-section goes up much faster than the fertile atoms capture cross-section as the neutrons slow down. Check out http://www.nndc.bnl.gov/sigma/index.jsp ... .0&nsub=10 . Select Uranium, then 233, then cross-sections (n,total fission) hit plot, then add to plot cart. Go back to the periodic table page (for me that means closing the plot window) and repeat for thorium, 232, (n,gamma) (you may have to scroll down for this), plot, add to plot cart. Go back to periodic table and select plot cart. You can see how above 1keV the cross-section ratio of fissile to fertile is around 7. Below 10eV the ratio is around 100. So if the edges have faster neutrons then the chances are that it is the fertile which will capture the neutron, while in the center where the neutrons are slower it is the fissile that will do so.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 192 posts ]  Go to page Previous  1 ... 9, 10, 11, 12, 13

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group