|
Post by eclipse on Feb 10, 2014 19:53:30 GMT 9.5
How would this compare to an IFR, especially with regards to safety, and half-life of final waste? Also, this doesn't seem very long to me: "By comparison, the TAP reactor can use current known uranium reserves to supply fully 100% of the world’s electricity needs for about 4,000 years." Barry once mentioned something about 50,000 years, but I think that was including thorium. nextbigfuture.com/2014/02/transatomic-power-molten-salt-nuclear.html
|
|
|
Post by David B. Benson on Feb 11, 2014 9:21:53 GMT 9.5
If I am correct in understanding that no pyroprocessing (electromechanical processing) is required this is a significantly better concept than a fast neutron reactor. Building a research version certainly seems to be a good idea, but I surely don't know enough to make final judgements.
Both designs are walk away safe.
The actual waste stream is similar to that from a fast neutron reactor. In both cases the percentage of actinides escaping into the waste stream is quite low so isolation for about 300 years suffices. After that the waste is no more radioactive than natural uranium (which has been used as a ceramic glaze, for example).
|
|
|
Post by eclipse on Feb 11, 2014 10:59:45 GMT 9.5
Significantly *better*? Wow. High praise indeed compared to all the wonderful aspects of the IFR.
|
|
|
Post by David B. Benson on Feb 11, 2014 13:58:14 GMT 9.5
eclipse --- Pyroprocessing is sufficiently expensive than fast neutron reactors cannot compete on an LCOE basis with light water designs. Possibly the TAP design does not require any comparable step. If that is the case then the LCOE appears to be quite a bit less than that for a new Gen III light water reactor.
Notice how heavily I've qualified...
|
|
|
Post by eclipse on Feb 11, 2014 14:28:25 GMT 9.5
eclipse --- Pyroprocessing is sufficiently expensive than fast neutron reactors cannot compete on an LCOE basis with light water designs. Possibly the TAP design does not require any comparable step. If that is the case then the LCOE appears to be quite a bit less than that for a new Gen III light water reactor. Notice how heavily I've qualified... I hear you. For now it's wait and see, but as a concept it's good.
|
|
|
Post by Ed Leaver on Feb 12, 2014 4:40:50 GMT 9.5
It is a concept. But bear in mind that while TAP's is a faster reactor, it is not a fast reactor. TAP try to optimize the neutron spectrum to give efficient thermal absorption by U235 and P239, while sufficient faster neutrons to split most of the remaining actinides, without having so dense a fast flux as to impose undue radiation stress on core components. But starting from just uranium, TAP still requires 1.8% enrichment, or 2.5x over the natural uranium which fast reactors can burn (and depleted uranium as well). This is a good fit with other estimates -- see "Sustainability" link below) of a 10ky natural uranium supply for fast reactors - thorium will run us extra by a factor of three or four. TAP is still far more efficient than LWR's, and may be a more economic (hence faster) path to SNF utilization and disposal than IFR. And SNF is a key public acceptance issue. But we'll see. S-PRISM is ready to deploy and is being flogged (with over-cooked pasta) by GEH. TAP is what, some minor start-up spun-off from a private uni? Pfffffft! </irony> However, I do caution about buying into TAP's "ZOMG!!! Peak Uranium is nigh!!! :eek:" marketing fud. At least not before reading Uranium and Depleted Uranium and Supply of Uranium first, from which we see WNA's "80-year" estimate to be highly qualified. I've a brief synopsis at Sustainability: How long can uranium last?
|
|
|
Post by eclipse on Feb 12, 2014 6:07:28 GMT 9.5
Hi Ed, that looks like a well argued page, full of information, but it would be great if it were on a wordpress blog or something. It's the readability & design (I may be over sensitive to these things running a design studio at home). Most webhosts have one-click Wordpress installs, and wordpress has heaps of free online video tutorials. You can probably even install a practice version and try importing all your website material into WP for a test run?
|
|
|
Post by Greg Simpson on Feb 12, 2014 14:39:55 GMT 9.5
I would hope they could nail down the life of the zirconium hydride moderator a bit better without a demonstration reactor. I don't know that it would be economical if it only lasts four years.
|
|
|
Post by Ed Leaver on Feb 12, 2014 15:18:27 GMT 9.5
So design the demonstration/research reactor such that the control rods are readily replaceable, on a rotating basis. MSR's are not pressurized, so operation need not be compromised. Four years probably isn't a deal breaker but the thing with reactors is, the material science pretty much has to be done in the same neutron environment as the proposed production plant. If you can find that environment at an operating production/research reactor e.g. Shippingport fine. But the TAP neutron spectrum is harder than anything currently operating in the U.S. but not nearly as hard as Russia's BN-600 which is probably already booked years ahead for such purpose and might not give the TAP crew the environment they need anyway. Obligatory oblique link: Beloyarsk-4 BN-800 Criticality Real Soon Now.
|
|
|
Post by eclipse on Feb 12, 2014 15:50:05 GMT 9.5
The BN-800's a breeder, but not an IFR... so it eats waste but ...? What's the key difference?
|
|
|
Post by jagdish on Feb 12, 2014 18:15:43 GMT 9.5
IFR means a breeder plus reprocessing. The Russians may be processing centrally at a few sites.
|
|
|
Post by Ed Leaver on Feb 13, 2014 3:11:19 GMT 9.5
To get near-complete fuel burnup a solid-fueled breeder must recycle its fuel rods multiple times to eliminate neutron poison decay products. The "integral" part of IFR means the recycling is co-located with the reactor, eliminating off-site shipment of partially burnt fuel seen by some as a proliferation hazard or terrorist target. Other designs e.g. SMR optimize long fuel resident times, meaning no refueling need take place for decades. The spent fuel should still be recycled for maximum burnup. Current liquid-metal fast reactors e.g. Russia's BN-600, BN-800, BN-1200 -- sorry folks, that's all we got -- employ uranium oxide and mixed oxide fuel elements similar to those in light water reactors. BN-1200 can use solid nitride as well. While these fuels have familiar production and materials properties, their reprocessing is more complex than the solid metal-fuel employed in EBR-II and S-PRISM, which was chosen for its relative ease of on-site reprocessing. (IFR solid metal fuels have some additional advantages as well, and a few drawbacks, relative to oxide fuels.) In contrast, liquid fuel reactors, be they LFTR or this TAP (currently) uranium design, do their fuel "reprocessing" (such as it is) on a continuous basis. Gaseous neutron poisons are removed by He sparging; most decay products are removed chemically e.g. precipitated as oxides and filtered on a continuous basis. See Section 2.6.3 "Waste Streams" and Table 1 page 16 of TAP Whitepaper. Molten salt reactors do have their advantages.
|
|
|
Post by eclipse on Feb 13, 2014 12:57:16 GMT 9.5
Thanks Ed. Does the reprocessing cost a lot more? I guess an 'energy park' with multiple reactors (all behind concrete walls!) would have one shared reprocessing unit to save on costs?
|
|
|
Post by Ed Leaver on Feb 13, 2014 17:05:12 GMT 9.5
I couldn't tell you on costs. Engineering economics is a hard science. But yes, S-PRISM farms appear to be intended for up to 5 two-reactor power blocks (@760mwe/block) fed by a common pyroprocessing fuel recycler. But there is considerable lag in recycling construction -- you don't build them till you need them, and in the early years there will LWR SNF reprocessors to be built to generate the LMR initial feedstock. This is outlined in Figures 5 and 10 of S-PRISM Fuel Cycle Study, Table III of which estimates (S-PRISM + processing) to be competitive cost-wise with LWR. Based upon 1997 numbers, ymmv. A more recent document is PRISM: A Competitive SMR, in which three 622 MWe two-core power blocks (6 reactors) surround a central Nuclear Fuel Recycling Center (NFRC) which in this scenario is designed to process LWR SNF and defense waste in addition to pyroprocessing the LMR fuel. Considerably more complex, but still part of the job that must be done. (Here and U.K. -- Australia presumably will be simpler.)
|
|
|
Post by David B. Benson on Feb 15, 2014 12:53:39 GMT 9.5
Ed Leaver --- Thank you for several matters, but particularly just now the link to the TAP Jan 2014 Technical Paper. All is clear except that this not-a-nuclear engineer does not know what helium sparging might be.
|
|
|
Post by Greg Simpson on Feb 15, 2014 13:37:33 GMT 9.5
|
|
|
Post by David B. Benson on Feb 15, 2014 13:40:29 GMT 9.5
Greg Simpson --- Thank you. Would argon sparging work about as well for this application?
|
|
|
Post by Greg Simpson on Feb 15, 2014 13:50:56 GMT 9.5
I'm certainly no expert. Argon sound plausible chemically, but could it be transformed by neutrons?
What is wrong with helium, anyhow?
|
|
|
Post by David B. Benson on Feb 17, 2014 9:30:01 GMT 9.5
Greg Simpson --- Helium is in short supply. Argon is readily obtainable from the air.
Don't know about neutron properties of either.
|
|
|
Post by Greg Simpson on Feb 17, 2014 14:26:40 GMT 9.5
We still have enough helium to fill giant blimps. While that may be unsustainable, the amount used for sparging should be relatively trivial. This is especially so if it can be recovered for reuse.
4He will absorb a neutron and then kick it right back out. Argon, and eventually almost everything else, will become radioactive. I'm unsure whether enough neutrons would be there for this to be a problem.
|
|
|
Post by David B. Benson on Feb 18, 2014 10:24:34 GMT 9.5
|
|
|
Post by jagdish on Feb 18, 2014 16:02:57 GMT 9.5
Russian BN-600 and Indian PFBR, a 500MW fast reactor are likely to be completed this year. Pyroprocessing, the processing suggested in IFR documents may follow later. No MSR is ready to be built for power production yet. Helium is in limited supply for most of the world. The only gas coolant in commercial use is CO2 in UK AGR's. Argon is vaguely discouraged. A gas cannot eat up too many neutrons, to my limited knowledge. Even nitrogen is talked about. It can be chemically active under certain conditions. Fast spectrum MSR's, if and when built, would be convenient for processing by chloride/fluoride volatility and electro-refining of used fuel. They would be useful for burning of used LWR fuel too. I was happy about TAP reactor assuming it was a fast MSR but am now doubtful about Zirconium Hydride moderator. Hydrogen and other materials eat up too many neutrons when moderated.
|
|
|
Post by David B. Benson on Feb 25, 2014 12:49:39 GMT 9.5
|
|
|
Post by Roger Clifton on Sept 25, 2016 9:30:40 GMT 9.5
DBB connected us to a Moltex presentation ppt wrapped in a pdf. The bullet-pointed, salesman's style of the presentation does not allow of credible analysis. They too, without explanation, refer to control by "a high temperature coefficient", apparently much higher than that described for the (solid fueled) IFR. I get the impression that a surge in reactivity only drops back when the volatiles (whose concentration varies) gasify and the liquid fuel foams over! Much more satisfying was the link to a recent (rev. July 2016) technical description of the TAP, also a two-fluid MSR reactor. TAP varies its moderators (ZrH) for control. The pdf describes fission product removal in the TAP as three processes -- sparging, filtration and "molten salt/liquid metal extraction". See its Table 1. "Sparging" with helium removes xenon, the most urgent FP to remove, but also tritium and krypton. "Filtration" removes colloids on a nickel gauze, while the nickel exchanges for the less reactive metals in the melt that might otherwise plate out. (This is the first reference I have seen to controlling radioactive scale in liquid fuel reactors.) The third appears to be a descendant of the batch electroprocessing used in the IFR and developed further since then in USA, France - and S Korea, where the fissiles are separated from the chloride melt into a liquid metal cathode.
|
|
|
Post by Greg Kaan on Sept 25, 2016 16:14:59 GMT 9.5
They too, without explanation, refer to control by "a high temperature coefficient", apparently much higher than that described for the (solid fueled) IFR. I get the impression that a surge in reactivity only drops back when the volatiles (whose concentration varies) gasify and the liquid fuel foams over! You seem to miss a fundamental point about MSR control - the large negative thermal coefficient prevents the situation you envisage. "Primary reactivity control is using the secondary coolant salt pump or circulation which changes the temperature of the fuel salt in the core, thus altering reactivity due to its strong negative reactivity coefficient." www.world-nuclear.org/information-library/current-and-future-generation/molten-salt-reactors.aspx
|
|
|
Post by Roger Clifton on Sept 25, 2016 19:24:37 GMT 9.5
Well, yes, I am skeptical of how well estimated is the DFK author's "strong negative temperature coefficient". In a slow neutron reactor, increased temperature of the fuel causes Doppler broadening of the low-energy resonances, with the implication that a slow neutron, solid fuel reactor in finely tuned balance, will tend to stay in that balance. If it gets too hot, fewer fissions occur and the reactor is stable. It is negative feedback at work. However the fuel temp coefficient varies greatly even in the lifetime of a solid-fuel core. In a fast neutron, solid fuel reactor – in finely tuned balance – the incoming energies of most of the neutrons are measured in hundreds of thousands of electron volts, so the Doppler shift (a fraction of an electronvolt) is much less effective. Instead, the structure of the core (solid fuel, in fine neutronic balance) expands slightly and an extra few neutrons escape, enough to collapse the balance back towards some equilibrium flux. "Plentiful Energy, Ch7" spells out the effect. The fuel of a fast neutron, liquid fuel reactor would also expand, and the reduced density of the fuel allow more flux to escape. However the logic that it would be enough relief assumes that the concentration of fissiles is constant, regardless of any surging of the pumps, or unclogging of the filters etc. Pumping coolant would give power control, but the delay for heat to conduct between the two liquids would make it too slow to control second-by-second transients. Anyway, how does the design maintain the fine tuning of these ( DFK, not TAP) liquid fuel reactors, which have no moderator or absorber in feedback loops? In a tube of standard diameter for an assumed standard concentration of fissiles? Hence my speculation that their claim to a particularly large negative temperature coefficient would have to rely on the fuel changing its density rapidly enough by forming gas bubbles from otherwise dissolved volatiles. Even then, would such a system respond fast enough for all contingencies?
|
|