Enrichment, or how I learned to stop worrying and love the SWU, part 4

Sorry it’s been so long since I last posted, but since I got on this topic I’ve had to “run the numbers” a fair number of times, and each time that I do that, I remember that we have computers that are a lot better at that sort of thing than I am, so after a little while I break down and start writing a code to “run the numbers” for me. And codes, as you probably know, are no smarter than the guy who writes them, so it ends up taking longer than I think it will.

Nevertheless, here’s a little code I wrote to go along with this series on enrichment. It runs in Java, and most people probably already have the stuff on their computers to make it work. It really shouldn’t matter if you’re running on a PC or a Mac or a Linux or Unix box. That’s the beauty of writing things in Java. And if for some reason you don’t have Java then you can pop over to java.sun.com to get it for free.

Anyways, the code incorporates the results of the equations that I presented in the last two parts of this series along with a few more that I may or may not get into tonight. The computational part of the code took me about five minutes to write. The pretty graphics and the user interface took much much longer. So I hope you like those. They’re there to make things easier for you at the expense of making them harder for me.


What the code shows is a little picture that looks like a Pac-Man eating some cheese. In the default case (isotopic separation of uranium), the Pac-Man represents the fraction of the feed that ends up in the tails. Which is most of it, which is why it looks like a Pac-Man. The thing that looks like a little cheese wedge is the fraction of the feed that ends up in the product. That part wasn’t all that hard to get to work, but what really took a while was the clever little graphics that show a pattern of dots in the midst of blue. Those dots are in proportion to how much of the enriched isotope is in the feed or the product. You can use the slider controls on the right-hand side of the program to change the enrichment values of the tails and product and you’ll see the density of those little dots go up and down in proportion.

I hope you like that. It took a while.

Anyway, you can see just by looking at Pac-Man and his little cheese that in a typical case of uranium enrichment (from natural levels to 3.5% or so) that most of the feed ends up in the tails, and a smaller fraction ends up in the product.


Let’s consider another case of enrichment—near and dear to my heart as an advocate for thorium and fluoride reactors: lithium. In the case of lithium, the isotope we want is lithium-7, which makes up more than 90% of natural lithium. But the problem with lithium is that we REALLY need to make sure that we get that pesky lithium-6 out of our mixture. Even a little bit of lithium-6 will make real trouble for a fluoride reactor. So when you switch the material over to lithium, the product slider in on a logarithmic scale, from 99% to 99.9999%. You’ll see that effect in the “dot density” on the picture and also in the amount of SWU’s required to reach that level of purity. As far as the tails enrichment of lithium, I’m not really sure what it should be, but I know what the product should be.

Another isotope of interest for future chloride reactors is chlorine. Chlorine-37 is the less common (~24% of natural chlorine) but more useful isotope for chloride reactors, and it is likely we’ll need to use high purity chlorine-37 in those reactors. Again, I’ve used a log-scale for the product from chlorine enrichment and a normal (linear) scale for the tails enrichment. For both lithium and chlorine I threw in some guesses as to what the feed and separative work might cost, but the reality is that I have no idea what those values should be. If anyone does know I’ll be more than happy to change the code to reflect those values.

Finally I looked at hydrogen enrichment in order to recover deuterium. This a really interesting case, because deuterium is such a tiny fraction of natural hydrogen that the feed-to-product ratio is CRAZY high. You have to go through a massive amount of hydrogen to recover a little tiny bit of deuterium. You won’t even see it on the graph—it’s so tiny. So it takes a lot of SWU’s and a lot of feed to recover deuterium, and deuterium is very valuable. A liter of heavy water (D2O) costs about $5000 the last time I checked.

Anyway, have a little fun with the program and I’ll tell you more about some of the other interesting aspects of enrichment that I have learned about in upcoming posts.

Comments

comments


12 Replies to "Enrichment, or how I learned to stop worrying and love the SWU, part 4"

  • Bill Woods
    August 16, 2010 (12:57 am)
    Reply

    "A liter of heavy water (D2O) costs about $5000 the last time I checked."

    Yikes, and only 4/20ths of that is deuterium. About $25k/kg?

  • Luke_UK
    August 16, 2010 (6:06 am)
    Reply

    Kirk, what Java VM are you using? I ask because at work (Win XP) I get:-
    java.lang.Exception
    at com.sun.javaws.Launcher.continueLaunch(Unknown Source)
    at com.sun.javaws.Launcher.handleApplicationDesc(Unknown Source)
    at com.sun.javaws.Launcher.handleLaunchFile(Unknown Source)
    at com.sun.javaws.Launcher.run(Unknown Source)
    at java.lang.Thread.run(Unknown Source)

    At home (Mac OS 10.5) I got the same, then upgraded java, and get complaints about incompatible class version numbers instead. You have better things to do than find workarounds for Sun/Microsoft/Apple bugs, but if there is a minimum version /system requirement, that would be useful to know.
    PS the brayton sim always worked fine, but now gives file not found.

  • Kirk Sorensen
    August 16, 2010 (7:07 am)
    Reply

    Hey Luke, give it another try. I recompiled it using an older version of Java and uploaded the resulting file to the server–I should have remembered–I've had this problem with Macs before and typically doing this has solved it. Give it a try and let me know.

  • David
    August 16, 2010 (9:31 pm)
    Reply

    Works great on my Mac after changing the Java 6 to be the top of the list. Very nice program and good information.

  • Nathan Wilson
    August 18, 2010 (10:24 pm)
    Reply

    Nifty tool. I was a little curious about HEU costs though…

  • Nathan Wilson
    August 18, 2010 (11:10 pm)
    Reply

    Hmm, in my quest to learn about HEU, I went over to http://www.fas.org/programs/ssp/nukes/nuclearcalc

    Of course the FAS's goal is make us worry more about enrichment. The point of their enrichment calculator is to show that making HEU is only slightly harder than making LEU; and it's convincing.

    So Kirk, is your point that we should stop worrying about U233, because U235 is so easy to separate? Ok, I'm slow, but I'm finally getting it (now that Iran has made LEU).

  • Kirk Sorensen
    August 18, 2010 (11:19 pm)
    Reply

    Actually I haven't gotten to the point of the whole series yet, which I'll be able to explain when I get to the point where I discuss the structure of enrichment cascades, and it's this–the arrangement of the cascades fixes the enrichment of the product and tails. A competent inspector could walk into an enrichment plant and identify quickly if the place is going to be making HEU or not.

  • Kirk Sorensen
    August 18, 2010 (11:23 pm)
    Reply

    Nathan, I tried the FAS calculator and couldn't get it to make any sense for me. Seems like they didn't spend much time in trying to make usable and intuitive sliders that bound the values appropriately.

    I have an updated version of the calculator with a log-scale on uranium enrichment that will let you get up to HEU values. I had bounded the previous version because I wanted good resolution at low enrichments, but I think I have now solved that problem.

  • Nathan Wilson
    August 21, 2010 (12:25 am)
    Reply

    Thanks Kirk. Yep, the FAS calculator is a little odd, you just keep entering values in different fields, then when it decides you have enough for a unique solution, out it pops.

  • Cyril R.
    August 21, 2010 (1:51 am)
    Reply

    Various CANDU PDFs on the internet give heavy water cost in volume at 300 – 1000 $/kg. I guess if you buy only one kg it will cost you a lot, though buying 50-100 tonnes for a reactor its a different story…

  • Bill Woods
    August 21, 2010 (11:10 pm)
    Reply

    The FAS thing doesn't work for me (on a Mac). Is it giving the same answers as the calculator on http://www.wise-uranium.org/nfcc.html ?
    WISE is an anti-nuke group, so I've been trusting them not to slant the numbers in favor of nuclear power.

  • Steve Koch
    September 1, 2010 (12:34 am)
    Reply

    Kirk,

    The java code worked great, love the HI!

    LFTR sounds great, what is a good web site that points out the problems of LFTR (it is always good to know both sides of the argument)?

    Was the bill by Joe Sestack to build a LFTR for a destroyer approved? What needs to be done from an engineering and regulatory perspective to make LFTR happen commercially in the USA? Is there a LFTR caucus in the house or senate? Why not push a bill to build a LFTR just to reduce the amount of nuclear waste? Is there anything organized about having us constituents talk to our congressman and senators about LFTR?


Leave a Reply