Energy From Thorium Discussion Forum

It is currently May 22, 2018 6:26 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 42 posts ]  Go to page Previous  1, 2, 3  Next
Author Message
 Post subject: Re: Protactinium-233
PostPosted: Apr 14, 2012 4:39 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
I understand and the idea is decent.

Normally the blanket is processed at a pace that would process the whole volume in about 6 months. Roughly then you would like to double the volume of blanket salt. Do a little calculation. Check to see if you get less Pa capture by processing the blanket salt in batches or by making the blanket itself twice as large with no external storage. It will take a bit of simple calculus. I'm pretty sure we do better with a blanket twice the size and continuous processing.

This also has two other advantages: a) thicker blanket to absorb more leaking neutrons and b) allows for smaller processing plant as it is working 24/7 so the throughput can be lower. For example, a 0.4m thick blanket surrounding a 2m radius, 4m tall right cylinder core is 36 m^3. Processing this in 6 months means a flow of 8.5 liters/hour. So the chemical processing plant can be quite modest.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 14, 2012 6:58 pm 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1950
Location: Montreal
Lars wrote:
.....a flow of 8.5 liters/hour. So the chemical processing plant can be quite modest.

I think so too ! ....a small processing plant like that shouldn't be extremely expensive either, as nuclear things go.

Now consider the possibilities for an MSR where we minimise the fuel salt volume.
5m^3 looks feasible for a ~1/2 GWe plant.
The whole lot could be processed in a couple of weeks, with benefits well beyond just saving Pa233.
(the main concern is that undesirable losses to the waste stream are sharply limited...)


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 15, 2012 7:21 am 
Offline

Joined: Apr 24, 2008 4:54 am
Posts: 491
Location: Columbia, SC
If you are going to do this you WANT a small processing plant. You do NOT want a big batch of Pa-233 to convert to U-233 all at once or over a short time frame. Ideally you want a conversion rate just fast enough to keep adding back to the fuel salt as more fuel is needed. If you wind up with a large inventory of fissionable material that is in storage waiting to be put back into the fuel salt, you have serious reactivity control concerns. Best to minimize those and breed just enough U-233 to add right back to the fuel salt.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 15, 2012 10:58 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
USPWR_SRO wrote:
If you are going to do this you WANT a small processing plant. You do NOT want a big batch of Pa-233 to convert to U-233 all at once or over a short time frame. Ideally you want a conversion rate just fast enough to keep adding back to the fuel salt as more fuel is needed. If you wind up with a large inventory of fissionable material that is in storage waiting to be put back into the fuel salt, you have serious reactivity control concerns. Best to minimize those and breed just enough U-233 to add right back to the fuel salt.

General point is right but the quantity of Pa233 present is constant as is how fast it converts to 233U. For an isobreeder at 1GWe it is roughly 74kg (1 tonne 232Th converted to 233Pa per year /365 days/yr * 27 days half-life). The only question is where it is stored.

But your general point is correct that you do not want to extract big batches of uranium all at once.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 15, 2012 2:12 pm 
Offline

Joined: Oct 06, 2011 1:34 pm
Posts: 81
Lars wrote:
I understand and the idea is decent.

Normally the blanket is processed at a pace that would process the whole volume in about 6 months. Roughly then you would like to double the volume of blanket salt. Do a little calculation. Check to see if you get less Pa capture by processing the blanket salt in batches or by making the blanket itself twice as large with no external storage. It will take a bit of simple calculus. I'm pretty sure we do better with a blanket twice the size and continuous processing.

This also has two other advantages: a) thicker blanket to absorb more leaking neutrons and b) allows for smaller processing plant as it is working 24/7 so the throughput can be lower. For example, a 0.4m thick blanket surrounding a 2m radius, 4m tall right cylinder core is 36 m^3. Processing this in 6 months means a flow of 8.5 liters/hour. So the chemical processing plant can be quite modest.


By "processing" do you mean removing Pa-233 from the blanket salt, or simply removing U-233? My idea is to remove Pa-233 mechanically from the blanket and let it decay to U-233 where it will not absorb neutrons from the core. At the same time, replace the blanket salt with fresh thoriated salt containing virtually no Pa-233.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 9:51 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
Pa separation is a different idea than I thought you were going for.
Pa separation is harder than uranium separation.
So I guess my first question would be is the Pa separation going to cost you more or less than storing the blanket salt and doing uranium separation?

Second, even if you do Pa separation, I'm pretty sure you will find that you are better off doing it continuously rather than in batches. The processing rate you would need to support would be lower.

Third, you will still need to do uranium separation for the blanket salt. No matter how fast you process it to remove the Pa some Pa will decay in the blanket before you get to remove it. So the uranium will build up over time unless you remove it (or the Pa separation process also removes uranium).

You haven't mentioned your technique for separating Pa.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 2:33 pm 
Offline

Joined: Oct 06, 2011 1:34 pm
Posts: 81
Lars, my technique for removing Pa-233 from the blanket is to pump the blanket salt out into a holding tank. There is no chemical separation. As the blanket salt is pumped out at one end, fresh thoriated salt is pumped in at the other end. Give the salt in the tank enough time for most of the Pa-233 to decay to U-233, then remove the U-233 hexafluoride gas, convert it with hydrogen back to uranium tetrafluroide, and inject it into the core. Cycle the residual thorium tetrafluoride in the holding tank back into the blanket. This process I envision could occur periodically, say every 120 days.

Because the "flow" can be regarded as pulsed with a long period, I think we could get the same net effect if the flow were continuous, at a rate that would replace the entire blanket salt every 120 days. It occurred to me that with continuous flow, the Pa-233 could be given more time to decay to U-233 if the quantity of salt in the holding tank were greater than the quantity of salt in the blanket, say a 3:1 ratio. In that case, the flow could be increased to a rate that would replace the entire blanket salt in a shorter period, say 90 days. The salt in the holding tank would have an average dwell time of 3 * 90 = 270 days, enough time for practically all of the Pa-233 to decay to U-233.

I have a question about the fluorination of uranium tetrafluoride to uranium hexafluoride gas. Does this occur spontaneously, or does it require the injection of fluorine or some other substance into the salt?


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 4:13 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
Steve Brown wrote:
Lars, my technique for removing Pa-233 from the blanket is to pump the blanket salt out into a holding tank. There is no chemical separation. As the blanket salt is pumped out at one end, fresh thoriated salt is pumped in at the other end. Give the salt in the tank enough time for most of the Pa-233 to decay to U-233, then remove the U-233 hexafluoride gas, convert it with hydrogen back to uranium tetrafluroide, and inject it into the core. Cycle the residual thorium tetrafluoride in the holding tank back into the blanket. This process I envision could occur periodically, say every 120 days.

Back to what I thought you meant in the first place then.
Quote:
Because the "flow" can be regarded as pulsed with a long period, I think we could get the same net effect if the flow were continuous, at a rate that would replace the entire blanket salt every 120 days. It occurred to me that with continuous flow, the Pa-233 could be given more time to decay to U-233 if the quantity of salt in the holding tank were greater than the quantity of salt in the blanket, say a 3:1 ratio. In that case, the flow could be increased to a rate that would replace the entire blanket salt in a shorter period, say 90 days. The salt in the holding tank would have an average dwell time of 3 * 90 = 270 days, enough time for practically all of the Pa-233 to decay to U-233.

But the flip side is the with the Pa in the blanket salt for 90 days you are running pretty close to not removing the blanket salt at all. The Pa half-life is 27 days so most of the Pa will decay in the blanket not your holding tank. Again though, take a look at the capture if you simply make the blanket larger rather than building a separate holding tank and pumping back and forth. I think you will discover that a larger blanket actually performs slightly better (in particular if you include the effect of lowering the neutron escapes). The cost limiter here is the cost of the blanket salt. For a smallish blanket it runs around $20M for a 1GWe reactor. Doubling it would cut the Pa capture losses in half and add $20M to the capital cost of the reactor. Doubling it again would cut the remaining Pa capture losses in half (saving 1/4 of the original Pa captures) at a further added cost of $40M. Clearly a cost versus performance tradeoff here.
Quote:
I have a question about the fluorination of uranium tetrafluoride to uranium hexafluoride gas. Does this occur spontaneously, or does it require the injection of fluorine or some other substance into the salt?

No thankfully it does not happen spontaneously. The last thing we want to happen is for us to have gaseous uranium floating around in our reactor. One must add fluorine to do this, typically a flame reaction. See the attached for a description of typical processes used in industry.


Attachments:
UF4 to UF6.pdf [1.66 MiB]
Downloaded 153 times
Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 4:36 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5058
The cost of that blanket is mostly in the 7LiF right? I think you can swap it for NaF if you really want a lot of blanket salt. The increased neutron captures shouldn't be as bad for a blanket salt as the core salt with the lower flux, right? You can also eliminate most of the tritium in the blanket this way. The higher melting point isn't a problem, since the lower heat generation in the blanket means a lower delta T in the blanket heat exchanger is required.

Even the $40-60 million for the LiF doesn't sound so much, though.

Simpler is better. Pa seperation is only attractive when rapid, but that's just where it's not clear whether its technically and economically feasible. Just go bigger on the blanket and leave the Pa there. You could also install a small passive heat exchanger that exchanges decay heat to the blanket salt to soak up some initial decay heat, while allowing a small primary fuel salt inventory.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 4:55 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
I'm not concerned about Pa capture in the blanket. I think the rationale for using 7Li in the blanket was so that there is almost no consequence to blanket and fuel salt getting mixed through a leak or in the chemical processing plant.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 5:39 pm 
Offline

Joined: Oct 06, 2011 1:34 pm
Posts: 81
Lars wrote:
But the flip side is the with the Pa in the blanket salt for 90 days you are running pretty close to not removing the blanket salt at all. The Pa half-life is 27 days so most of the Pa will decay in the blanket not your holding tank.


My original idea was to flush the blanket salt out to a holding tank when the Pa to Th ratio exceeds a certain value, and refill the blanket with salt containing Th and virtually no Pa. For a Pa to Th ratio of 1:3, I don't know how many days that would take. It would depend on the neutron flux and other factors. It could be considerably less than 90 days, a number I picked to illustrate the concept, which is to minimize neutron capture by Pa-233, by flushing it out of the blanket at a rate great enough to increase significantly the probability of neutrons hitting Th instead of Pa in the blanket. There is virtually zero probability of Pa-233 capturing neutrons from the core while in the holding tank. If instead we simply increase the mass of the blanket, all of the Pa-233 remains in the blanket, where it can capture neutrons, before it decays to U-233.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 8:10 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
Steve Brown wrote:
Lars wrote:
But the flip side is the with the Pa in the blanket salt for 90 days you are running pretty close to not removing the blanket salt at all. The Pa half-life is 27 days so most of the Pa will decay in the blanket not your holding tank.


My original idea was to flush the blanket salt out to a holding tank when the Pa to Th ratio exceeds a certain value, and refill the blanket with salt containing Th and virtually no Pa. For a Pa to Th ratio of 1:3, I don't know how many days that would take. It would depend on the neutron flux and other factors. It could be considerably less than 90 days, a number I picked to illustrate the concept, which is to minimize neutron capture by Pa-233, by flushing it out of the blanket at a rate great enough to increase significantly the probability of neutrons hitting Th instead of Pa in the blanket. There is virtually zero probability of Pa-233 capturing neutrons from the core while in the holding tank. If instead we simply increase the mass of the blanket, all of the Pa-233 remains in the blanket, where it can capture neutrons, before it decays to U-233.

You need to work the numbers. The ratio of Pa to thorium in the blanket will be more like 75kg / 40 tonnes. If you make the blanket twice as large then the ratio would be 75kg to 80 tonnes. Yes you can move some of the Pa into the holding tank but at the same time you move the thorium with it.
Given a certain size blanket adding a holding tank will decrease the Pa capture at the cost of buying more blanket salt. I think you will find that the faster you pump blanket salt through the holding tank the better you improve performance. But even better would be to use that extra blanket salt you bought and simply make the blanket larger.

I don't know your background but if you've had an engineering/math/physics/chemistry education then you likely have taken a differential equations class. This is a simple rate equation problem.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 9:38 pm 
Offline

Joined: Oct 06, 2011 1:34 pm
Posts: 81
Lars wrote:
...The ratio of Pa to thorium in the blanket will be more like 75kg / 40 tonnes....


Really? The ratio 75:40,000 reduces to 1:533, or 1 part Pa to 533 parts Th in the banket salt, even after extended running and equilibrium is attained. That makes neutron capture by Pa-233 insignificant, not worthy of any special design features to make it less. Are you sure that 1:533 ratio is correct, or approximately so?

My fields of expertise are computer science and electronics. When it comes to nuclear reactor design, I'm an interested observer.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 16, 2012 10:00 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3068
Equilibrium is attained pretty quickly since the half-life is only 27 days. Within six months you are at equilibrium. Pa capture is an issue to folks using thermal designs and where they are trying to achieve breeding. In particular for the 1.5 or single fluid designs where most of the Pa is in the core with a higher flux and where the amount of thorium you add has to be matched with fissile (expensive). People who are interested in optimizing breeding are typically trying hard to minimize the fissile inventory.

Pa capture in the blanket is generally not a significant issue. It is easy to increase the blanket size. The amount of Pa is rather fixed by physics. The amount of thorium is a design choice. But thorium is cheap and the more thorium in the blanket generally the better.

MSBR is the most developed large scale LFTR plan. It contains 68 tonnes of thorium. (It happens to be a single fluid reactor so all the thorium is in the core). The cross-section of Pa is about 4x that of thorium so the effective ratio is smaller than you think but still the point is still the same.


Top
 Profile  
 
 Post subject: Re: Protactinium-233
PostPosted: Apr 17, 2012 7:51 pm 
Offline

Joined: Apr 24, 2008 4:54 am
Posts: 491
Location: Columbia, SC
Steve Brown wrote:
Lars wrote:
...The ratio of Pa to thorium in the blanket will be more like 75kg / 40 tonnes....


Really? The ratio 75:40,000 reduces to 1:533, or 1 part Pa to 533 parts Th in the banket salt, even after extended running and equilibrium is attained. That makes neutron capture by Pa-233 insignificant, not worthy of any special design features to make it less. Are you sure that 1:533 ratio is correct, or approximately so?

My fields of expertise are computer science and electronics. When it comes to nuclear reactor design, I'm an interested observer.


Be careful using a simple ratio, neutron cross sections for absorbtion will make a big difference.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 42 posts ]  Go to page Previous  1, 2, 3  Next

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