What's in Spent Nuclear Fuel? (after 20 yrs)
Lately I’ve been looking a lot at spent nuclear fuel, particularly what’s in it and what’s radioactive after a while.
I’ve seen graphs of fission product distributions before, but they’re always of fission products by atomic mass, and they’re usually showing the distribution right after the fission event.
I wanted to know what was in there from an elemental perspective, because if you’re going to do chemical processes on the spent nuclear fuel, you’ll be removing elements, not isotopes. So how do they all rack-and-stack overall? Well, here’s the results for a typical light-water-reactor fuel element, irradiated for a modest 400 days and then allowed to cool down for 20 years. Everything is scaled to one metric tonne of uranium before irradiation, and values for masses are given in grams. So xenon, for instance, constitutes 0.13% of the mass of the original fuel, per tonne of uranium.
Some of the smaller concentrations of fission products are easier to see in a log-scale graph of the same data:
Here’s another image that might help, that shows how the fission products “grow in” to the fuel as it is irradiated. Note that this is a 3 year irradiation, so there’s more of everything in this distribution.
As you can see from the graph, there’s not ALL that many significant (from a perspective of mass) fission products in the spent fuel. There’s xenon (#54) and neodymium (#60). Then there’s zirconium (#40) and molybdenum (#42) and ruthenium (#44). Cesium (#55), barium (#56), lanthanum (#57), cerium (#58), and praseodymium (#59) all figure in at varying levels of importance. And there’s samarium (#62) in there to make things difficult.
But it’s a smaller list than I would have thought, and the xenon, neodymium, molybdenum, and lanthanum are all recoverable at this stage. Something to think about–even the fission products of spent nuclear fuel probably aren’t really “waste” either.
Elements 44 through 46 are platinum group metals. Here are the latest spot prices.
Johnson Matthey Base Price $/oz (31.1 grams) http://www.platinum.matthey.com/
46 Pd – 484
45 Rh – 2425
44 Ru – 240
There might be $10,000 worth of metals in a ton of spent fuel. How many natural ores are that rich?
http://www.popsci.com/science/gallery/2010-05/gal…
Neodymium — 60
1 kg/T
Reserves: Unknown but scarce
Cost: $28 per pound
Critical for: Electric motors
Outlook: Poor
This irreplaceable rare-earth metal is used to make light, powerful magnets in motors in cars, hard drives and wind turbines. Lack of Neodymium could be a limiting factor in the adoption of green technology, such as gearless wind turbines. It’s also used in military lasers for targeting and communications. Even though there are about 100 million tons of it in reserves, rising demand means the U.S. needs to scout out more deposits and improve extraction techniques for mixed ores.
Tellurium 52
100 g/T
Reserves: 22,000 tons
Cost: $66 per pound
Critical for: Solar cells
Outlook:Fair
Almost all tellurium is found as a trace element in copper ore, and very little can be recycled. No other element turns sunlight into electricity as well, and solar cells using tellurium semiconductors are inexpensive. If the tech proliferates, manufacturers might have to move on to less-efficient or pricier elements, such as selenium.
Thanks Kirk,
One of the ASME's senior editors has invited me to write an article about a "new" way of turning the sorts of halide salt wastes generated by breeder-type reactors (both LFTR and IFRs)into a glass "waste form" & this stuff will be very useful.
Is your data available in an EXCEL file?
Excellent resource page here. I will be referring to it often.
Keep in mind that the market price and availability of lesser used elements is somewhat complex. Most importantly these prices are only somewhat supply & demand driven. In general one can count on all sorts of manipulation going on in the way of price support. Thus it would be unwise to assume that the prices represent any firm potential for making money by exploiting used nuclear fuel.
Hey, why is removal of samarium so difficult from a LFTR? Can we not just fluoridate the 233-U, purify the salt*, reduce the 233-U and reintroduce the pair?
* Hah! There's my "???, profit!" I'm basically asking if, after isolating the 233-U from the salt by turning it to UF6, is it very difficult to pull all of the fission products out of solution? Couldn't we borrow pyroprocessing techniques (uses LiF/BeF2 as the solvent anyway) or centrifugal separation to purify the salt?
I've always doubted the economics of trying to recover fission products. However, does anyone know the isotopic distribution of Xenon from either LWRs or LFTR/MSRs? There may be a short term but large demand for the isotope Xe136 (naturally about 9%). My university has a large particle physics group and are part of a team designing neutrino experiments that need Xenon enriched in 136 (they have basically taken everything Russia can provide just for a small scale experiment).
http://www-project.slac.stanford.edu/exo/
David L.
6.66% fission yield for 136Xe
21.5% fission yield for all stable Xe's.
So 30% Xe136.
Radioactive Xe isotopes decay quickly (131mXe has a 12 day half-life) so within 6 months the Xe is stable. We can isolate it by fractional distillation using liquid nitrogen.
This is for u233 but I bet the numbers are pretty similar for u235.
I exchanged emails with someone on the Xenon project. Their goal (if funded) is a machine using 10 tonnes of enriched Xenon enriched to 80 to 90% 136 (at least a 100 tonnes natural). With the entire world's production about 40 tonnes a year this is a tall order!
Lars (or anyone), are there other fission yields that decay to stable Xenon (ideally 136) or do those percentages include things that reasonably quickly turn to Xenon. I wonder what Le Hague does with their Xenon?
David L.
Addendum. In an LWR most of the Xe135 gets converted to Xe136 so you can add 90% of 4.8% yield giving you
11% yield for 136Xe
26% yield total stable Xe
or 42% 136Xe.
(Assuming no neutron captures in the stable Xe's – not a fair assumption but I'll let someone else fuss over this effect).
As a propenent of LFTR, I think the most important number in this is that after 3 years in an LWR, 95% of the "fuel" has not been touched!
Hm.
People seem to think this article is more about isolation of fission products for sale than purifying the fuel salt for uninterrupted energy production.
I feel like FP isolation doesn't need to be a part of the LFTR core for the reactor to be considered complete; I'm not saying there's no room for it at an auxiliary plant, but the reactor's only real concern is that the more troublesome FPs get pulled out.
Maybe I'm just lazy, but if I'm designing a reactor, FP isolation just seems like a "not my problem; hire someone to design a secondary system" issue. I'm just worried about fuel purification.
This post reminds me of earlier posts, including one that had the spent uranium graphic, about fluoridating waste to reduce its volume and separate uranium for re-enrichment. I think that it would be simpler to extract U-Pu to create MOX fuel for CANDU reactors and just give the rods to Canada. The waste volume reduction could justify the cost. Recovering useful elements is a bonus.
BTW, I thought online reprocessing was a major part of the molten salt reactor concept. Regularly pull out the U (and TRU, depending on the design) and put it back in the reactor along with fresh fuel. Separating the FPs is a different problem
We want to pull out the fission products to reduce the number of neutrons they steal. All LFTR designs use He sparging to pull out the krypton and xenon (typically w/in a minute). The noble metals will come out whether you plan for it or not so one should plan for it so they don't plate out on the cold end of the heat exchanger and cause a decay heat problem during an emergency (typically w/in hours). Finally you have the salt seekers. Designs vary in the extraction time for these from months to decades.
The discussion of separation of fission products is in a secondary processing system. This also relieves the local power plant of any requirement to deal with americum or curium and even plutonium. Presuming a secondary processing systems also enables designs that extract neptunium.
The main problem with fuel separation and regeneration isn't technical (though it might be economically unfeasible of course) but legal.
Most countries (maybe even international treaties like the NPT) ban the refining of spent nuclear fuel for fears that it might produce quantities of weapons grade material.
This fear was driven by the anti-nuclear lobby as far back as the 1970s and has never been corrected.
It has cause fast breeder reactors to be shut down while still under construction (research Kalkar for a prime example).
I've done some nuclear waste related research as part of graduate work in the 1990s, developing new non-destructive methods of classifying low grade waste as nuclear or non-nuclear for disposal purposes.
These laws were the main reason for that research, it was legally impossible to even sample the waste material for chemical analysis of its content. All had to be determined solely from analysing emitted radiation (or in case of the method I used, scattered radiation from bombarding the samples with a known source).
If nuclear waste material can be refined and broken down into different category batches, storage requirements (and duration) go down drastically, removing the main remaining problem many have with nuclear energy (no doubt this alone is a reason some people don't want such technology to be employed).
The Uranium, Plutonium, and some other elements can be reused for new fuel rods, other elements used for radiochemistry and medical uses, and the remaining nuclear material stored in smaller containers for shorter periods (it will be mainly extremely high radioactive waste, no more than a small bucket a year for an average reactor to be stored for a few decades at most).
I've since left the field, so don't know the current state of technology, but that's as things stood when I was engaged in it.
the make-up of waste as a function of time by Westinghouse is a good graphic. Does anyone have a similar one for thorium cycle?