|
Post by cyrilr on May 28, 2019 21:46:47 GMT 9.5
Ok, so here is how I see it.
1. Commercial enrichment facilities make low enriched uranium which can't be used to make bombs. Can the facilities be modified to produce weapons grade uranium? Yes, with some modifications - things like criticality, transport casks with absorbers and so on will need to be changed. The equipment needed for this is not present so will have to be brought in secretly. There are a limited number of commercial enrichment facilities (they serve many reactors) so this is easier to control. On the negative side, a U bomb is easier to make than a Pu bomb.
2. IFR equilibrium fuel has poor isotopic quality which can't be used to make bombs. Can the IFR be modified to produce weapons grade plutonium? Yes, with very few modifications in fact. Simply replace some of the fuel assemblies (near the edge) with depleted or natural uranium. This will create bomb grade plutonium in those assemblies. Since the equipment is present in the IFR - both fuel fab and recycling - this would be relatively easy to do. It would be very difficult to control 1000 IFRs versus a dozen enrichment plants. On the plus side, a Pu bomb is harder to make than a uranium bomb.
3. Neither 1. Nor 2. is particularly relevant to the threat of proliferation since the main methods to produce weapons grade materials are through small Pu production reactors or small dedicated enrichment facilities. Even easier would be to obtain an intact weapon from somewhere in the first place. Closing down the commercial powerplants would do little to reduce proliferation risks.
Proliferation minded folks tend to not want nuclear anything: anything with a lot of neutrons can be misused, therefore it is to be avoided. Nuclear power is a no-no to these people altogether. Proposing techno fixes doesn't help since the techno fix can be busted or avoided or misused in some other way by would be proliferators.
Only 3. has a major chance of success, because it makes 1. and 2. moot.
|
|
|
Post by engineerpoet on May 29, 2019 0:41:53 GMT 9.5
Can the IFR be modified to produce weapons grade plutonium? Yes, with very few modifications in fact. Simply replace some of the fuel assemblies (near the edge) with depleted or natural uranium. This will create bomb grade plutonium in those assemblies. Not quite; there is a lot more to making bomb-grade Pu than just irradiation. Not only do you have to keep the irradiation rather short (which requires lots of swapping blanket elements in and out—if that forces the reactor to be powered down, that will be very obvious) but you also have to separate the Pu from both the un-converted uranium and also from fission products. This requires wet chemistry like SOLVEX or PUREX. Except for 1 thing: the IFR's pyroprocessing technology is engineered to NOT produce pure Pu. The Pu product of the IFR is contaminated with an unavoidable fraction of fission products. It wouldn't take much to leave enough gamma-emitters to fry the explosives around a weapons "pit", the electronics to fire them and the people trying to put it all together. Except in a world of Integral Fast Reactors or the equivalent with less-distributed fuel reprocessing, nobody would have PUREX unless they were up to no good.
|
|
|
Post by engineerpoet on May 29, 2019 0:55:10 GMT 9.5
One more thought:
Even in a world of IFRs, the core composition can change the breeding ratio or even turn it into a burner. Anti-proliferation would be easy to arrange with just 2 changes:
1. Make the breeding ratio slightly sub-unity, requiring a bit of make-up fuel from time to time.
2. Limit the local reprocessing capacity to only what the reactor needs for its own operation.
Given those restrictions, any diversion of fuel, reprocessing or neutrons will cause the reactor to go sub-critical and shut down. That's another thing that would be easily detected, and would make the consumers of the reactor's electric power unhappy in the bargain.
As for having to engineer a centrifuge cascade to produce HEU, perhaps you're right... but the long, thin form factor of centrifuges is practically made for preventing criticality accidents, and there's not much to prevent the placement of cadmium or boron neutron absorbers around and between them.
|
|
|
Post by cyrilr on May 29, 2019 1:41:13 GMT 9.5
Irradiation time is of lower relevance, it is mainly a fluence thing - you want to keep fluence limited. The edge of the core receives much lower fluence than the central fuel assemblies. Typically a Gaussian distribution is seen unless special measures are taken (radial fissile zoning and so on). No matter how hard we try the flux always cliffs off at the edge of a core... in a power reactor we always try to flatten the flux as much as possible to get the most power out of the core but low flux at the edge is not possible to avoid. So for a given irradiation time, the edge of the core will see a lower fluence. Putting depleted uranium assemblies there - a "breeding blanket" will enhance this effect. DU assemblies make little power so encourage flux depletion in that area, and peaking in the central core.
It is also quite conceivable to dream up credible excuses for shutting down earlier - e.g. turbine maintenance. It is more of a question whether or not the plant in question is subject to safeguards or not.
In terms of the pyroprocessing: it is designed to keep Pu with the transplutonium elements and fission products, but their concentration is much lower for reduced fluence that would be seen at the edge of the core. One could have a quite low concentration of FPs and MAs by having a full radial blanket.
Fast reactors on Pu/U238 cycle have a lot of excess neutrons. These could be mis-used. Sub-unity breeding typically only happens with small fast reactors, but those have such an enormous fissile startup loading they don't look so attractive. Reactors have big economies of scale. It will make sense, in the long run, to have big fast reactors.
In terms of reprocessing capacity - I'm afraid we are talking about very small capacities here, it is basically lab quantities. It would not be too difficult to increase capacity. Nuclear breeders are very efficient, so material flows are very small.
As for UF6 - as a gas it won't go critical, even 90% enriched. The inside of the centrifuge train isn't the problem. Storing it as a solid is a different matter. F is a moderator so bomb grade UF6 needs special precautions to prevent it from going critical - can't just sublimate it into a drum or storage vessel, it needs special geometry and more likely strong absorbers like boron. Which is quite possible to do. Doing it legit is tough - storage and transport requirements need a very low k-eff. The usual containers for shipping LEU would not be ok for shipping HEU.
For the record I am not arguing IFRs are a big proliferation hazard. I am arguing that we need to be careful what we claim about proliferation. Proliferation is very much like the "waste problem" and "safety issues" with nuclear plants - really not a big deal, simply because the two main, proven methods of nuclear weapons material production don't require power reactors in the first place. This was Cohen's argument (later adopted convincingly by Bill Hannahan) and it is the most useful argument I've found so far.
|
|
|
Post by engineerpoet on May 29, 2019 7:48:14 GMT 9.5
Irradiation time is of lower relevance, it is mainly a fluence thing - you want to keep fluence limited. If I understand right, it would be ideal to have extremely high fluence for a brief period (close to a delta function) so as to not have time for the Np-239 to form before the irradiation ends. With Np-239 you get formation of Pu-240, and with Pu-239 you get fission 65% or more of the time. True, but anyone shutting down on a regular schedule for "turbine problems" would be waving a red flag saying "I'm doing something suspicious!" Yes, but your concentration of desirable Pu would also be quite low. You'd be reprocessing a large amount of material to extract the fraction of a percent of Pu. If you also have to reprocess to keep your reactor running and you only have enough capacity for the reactor, the reactor eventually stops running. This is where you can get clever. There is no need for the reactor to have a pure fast spectrum; epithermal reactors work just fine, and as a bonus they need much reduced fissile loadings. Just hit the sweet spot where the breeding ratio is almost exactly 1.00 and you've pretty much eliminated the possibility of diversion without having to worry about monitoring. The reprocessing systems at Hanford were anything but small. If you're trying to extract 20 kg of Pu from blanket rods at 0.5% or less, you've got to reprocess at least 4 tons of raw material that is already radioactive. You're going to need hot cells, remote manipulators, the works. It's going to take quite a bit of room, cost a lot of money... and if your reactor shuts down due to lack of fuel in its recycle stream, you've screwed yourself. I was thinking of liquid UF6 in tanks filled with boron carbide marbles or the like. Cadmium honeycomb would be great but you can't just take off a valve and pour a whole bunch of it into a storage vessel. What I'm arguing is that enrichment IS a big proliferation hazard, and we can make fast reactors which make proliferation so hard that we can let just about anyone have them. Make systems that don't NEED an army of watchers and we'll all be a lot happier.
|
|
|
Post by David B. Benson on May 29, 2019 11:29:32 GMT 9.5
I visited a plutonium production reactor at Hanford in the 1950s. One of a lengthy row.
Much later I taught at least two students who subsequently spent their entire career involved in the Hanford site cleanup, a still on-going event. And yes, some of the plutonium extraction halls are ultra massive and still there.
It takes a state actor. The most recent for plutonium is India, although I know nothing about their extraction facility. Of course Pakistan has been satisfied with uranium. Oh yes, then there is North Korea.
|
|
|
Post by cyrilr on May 29, 2019 16:29:57 GMT 9.5
I visited a plutonium production reactor at Hanford in the 1950s. One of a lengthy row. Much later I taught at least two students who subsequently spent their entire career involved in the Hanford site cleanup, a still on-going event. And yes, some of the plutonium extraction halls are ultra massive and still there. It takes a state actor. The most recent for plutonium is India, although I know nothing about their extraction facility. Of course Pakistan has been satisfied with uranium. Oh yes, then there is North Korea. We're not talking about large, centralized, aqeous processing here like Hanford. We're talking a small, modular, pyroprocessing facility co-located with an IFR. BIG difference. If every IFR requires a Hanford, the concept is dead on economic grounds, obviously!
|
|
|
Post by cyrilr on May 29, 2019 16:56:24 GMT 9.5
Pretty small effect - U239 has a half life of 23 minutes. Slightly bigger effect for Np239, with about a 2 day half life. Still pretty small. Most of the Pu240 will come from capture on Pu239.
Interestingly this effect is pretty significant for thorium reactors though, where a big protactinium decay reactivity effect occurs over a long period.
Probably not. Lots of problems usually occur with startup. Some plants keep getting the same problem over and over. The Japanese plants have developed a recurring case of condenseritis. Their capacity factors are relatively poor as a result.
Not much is needed. 7 kg would be enough. Fast reactors need a lot of fuel for startup, 4000-9000 kg/GWe.
Again this assumes the design and operations are controlled. Why not swap out the fuel assemblies and put in tighter pitch ones, faster spectrum? The means to make the fuel assemblies is present onsite for the IFR. If you are assuming controlled design, then how would you make bomb grade uranium from a commercial enrichment plant not designed to make or transport that?
Again, we are not talking about large centralized aqeous like Hanford, we are talking a small modular IFR with onsite pyroprocessing. BIG difference. 4 tons/year is 11 kg/day pyroprocessing - pretty small facility.
sure, technically this is all feasible, but it requires non standard equipment to be brought in a high security, highly bureaucratic enrichment plant. If you assume that is done, why not assume simple modifications to an IFR to make Pu239 from DU FAs?
As I tried to point out, the problem with this argument is that power reactors don't run on bomb grade uranium, they run on low enriched which can't be used to make a bomb. So it is a question of modifications - can modifications be made to turn an enrichment plant into a bomb grade materials producer? Yes. Can modifications be made to turn an IFR into bomb grade materials producer? Yes. In many ways the IFR modifications are much easier, since the equipment is already present on site - fuel fab, pyroprocessing.
Another problem with your argument is you're condemning all existing reactors, so what, replace with coal? IFR can't be deployed right now, still needs work on the pyroprocessing, and even if it could it would be no good to replace LWRs with IFRs when you could be replacing coal plants with IFRs. As long as we use coal this reactor favoritism is a spoilt rich kid attitude.
So I have to say your argument isn't very convincing and not strategic or political. Much better to call a spade a spade - proliferation isn't a valid argument against nuclear power, because the 2 main, proven methods of making bombs don't require power reactors.
|
|
|
Post by engineerpoet on Jun 4, 2019 6:39:06 GMT 9.5
this effect is pretty significant for thorium reactors though, where a big protactinium decay reactivity effect occurs over a long period. That's actually what I was thinking about. Not just the neutron-capture cross-section is high, but the decay energy is considerable. I calculated the decay heat as a percentage of full-power operation in a unity-ratio breeder and it came to several percent, enough to make a serious contribution to decay-heat removal requirements. I was pondering the possibility of a LWBE-type core in a NuScale and concluded that the water pool would have to be increased quite a bit to achieve passive shutdown. But how many of them keep doing it on a schedule you can mark off on a calendar before the fact? This is why you go epithermal for the mass-market version. The fissile loading is radically reduced, and if the spectrum and core design hit 1.000 breeding ratio anyone trying to divert either fissiles or neutrons is stuck with shutdown now or softening the spectrum further, cutting the breeding ratio to sub-unity and shutdown later; either one leaves tracks that indicate what was done. If the isotope mix is sufficiently far from weapons-grade to make the material militarily useless, you've effectively closed off all simple proliferation paths. It's not necessary to make it impossible, just enough more difficult than uranium enrichment that nobody will bother. That requires a higher fissile loading. Where do they get the fissiles? Chicken/egg. It's inherent in any system which can begin with 0.7% U-235. You yourself said that criticality is not an issue in gaseous UF6, and if I can design a small-diameter tubular tank set surrounded by boron or cadmium and connected by a manifold, any actual student of nuclear engineering ought to be able to do it in their sleep. But how much can you divert without forcing the reactor to shut down? Legitimate problems with the pyroprocessing system would leave an inventory of fuel needing to be reprocessed. If you've gone to an epithermal spectrum which produces a 1.000 breeding ratio, using DU breeding assemblies to make Pu for diversion takes neutrons the reactor needs to keep itself going. Shortly after you do this, the missing fissiles force the reactor to shut down. You can't harden the spectrum without adding fissiles to make up for the decreased fission cross-section. Checkmate. Per Barry Brook, an AP1000 takes 20.8 tons/year of fuel at 4.45% enrichment. That requires about 210000 kg of natural uranium at 0.711% U-235 with tails at 0.3% and 127598 SWU per year. That leaves 630 kg of U-235 in the tails every year, when the critical mass is about 52 kg. Reducing the tails concentration to 0.2% cuts the feedstock requirement to about 173000 kg and increases enrichment burden to 157563 SWU, about 30k SWU more. Converting that saved 37000 kg of NU to bomb-grade at 90% U-235 and the same 0.2% tails produces about 208 kg worth and takes about 47280 SWU. So, given the enrichment and conversion parts of the fuel cycle, it would be possible to produce about 2 Little Boy equivalents per year from the uranium feed required for an AP1000 at the expense of about 77000 SWU, consuming less than 4 GWh/year (about 440 kW average). It would be VERY easy to hide this level of effort given the hardware and the desire to use it for those ends. Wrong. Pyroprocessing leaves too much in the way of fission products in the Pu fraction to allow weapons manufacture even if the isotope composition is good enough. At least one additional step, such as PUREX, would be required. There are three further requirements for proliferation: 1. The reactor has to produce excess fissiles (which can be prevented by selecting the spectrum for unity breeding ratio). 2. At least a fraction of the reactor's bred fissiles must be produced under conditions which yield weapons-grade isotope ratios. 3. There must be sufficient reprocessing capacity to not only satisfy the reactor's own need for replacement fissiles, but also handle the diverted weapons-grade stream. As I noted above, the whole question can be rendered moot by using an epithermal neutron spectrum which yields a unity breeding ratio. The epithermal spectrum reduces the required fissile load well below the needs of a fast-breeder, and produces no excess fissiles. A reprocessing system strictly designed to satisfy the reactor and no more would be fully subscribed just to keep up with replacement fuel demands and have no excess capacity to produce weapons. Last, even if some material was irradiated under conditions to yield a weapons-grade isotope mix, (a) this would consume neutrons required to replace the reactor's own fissiles and lead to the reactor shutting down and (b) pyroprocessing would not yield a sufficiently pure Pu product to make weapons; other processing would be required. What's easier, satisfying all those conditions or finding less than half a megawatt to feed some centrifuges working overtime? I'm condemning NOTHING. I'm recognizing a truth about nuclear proliferation, which stands in the way of expanding nuclear energy via LWRs. That truth is true no matter what I say or fail to say. That same truth is recognized and codified in anti-proliferation treaties, and all I am doing is acknowledging that FACT. You need to face that FACT also. Right now, LWR fuel production is confined (save for Iran) to a small number of states which already have nuclear weapons and present no proliferation hazard. We don't need to do anything about the current LWR fleet; they will eventually reach retirement age no matter what else happens. However, if we're going to deal with the twin problems of climate change and energy insecurity due to fossil-fuel depletion, we need a substitute which is far more abundant and clean. Trying to increase world uranium production to 650k-odd tons per year, which is what LWRs would require, probably won't fly even if we could supervise the enrichment systems well enough to avoid more nuclear weapons breakouts. S-PRISMs and the like can do the job, though, and would only consume about 5000 tons per year worldwide. Starting a Fermi 1-class reactor requires only a modest amount of feedstock. Fermi 1's core contained 484 kg U-235 at 25.6% enrichment, for a total of 1891 kgHM in the core. My enrichment spreadsheet shows that this could be made from just over 60 MTU at 0.2% tails, requiring a bit more than 90k SWU. If uranium mining could be increased by 30k tpy, it would be sufficient to start 500 Fermi-class reactors at 200 MW(t) (probably 80 MW(e)) apiece every year. That's a ramp rate of 40 GW(e) per year. That would replace the US LWR fleet in 2.5 years, and all of net US electric generation in just over 11 years. The DU tails from the enrichment of the startup fuel load would be sufficient to run the reactor for its own lifetime, and quite a few more. It convinced the PTB before I was born. Are you even bothering to read what I write? That's what I've been saying: uranium enrichment (which is REQUIRED for the LWR fuel cycle) is sufficient for a nuclear weapons program. Therefore, scaling up nuclear energy using LWRs is problematic for anti-proliferation efforts. We would be better off with some combination of FBRs and epithermal reactors using a fuel cycle such as pyroprocessing which produces a recycled fuel product which is both isotopically unusable for weapons and too contaminated with "hot" fission products to fabricate into a viable bomb even if it was.
|
|
|
Post by cyrilr on Jun 8, 2019 19:28:13 GMT 9.5
Sorry EP you are missing the main point completely, once again. I am tired of repeating myself.
|
|
|
Post by cyrilr on Jun 8, 2019 20:46:31 GMT 9.5
"But how many of them keep doing it on a schedule you can mark off on a calendar before the fact?" Doesn't matter. I can tell you a hundred different ways to hide this up, from sabotaging some of the plant's systems, forcing to run on part load (even 99% load would allow enough reactivity savings to make weapons) just one example. "This is why you go epithermal for the mass-market version. The fissile loading is radically reduced, and if the spectrum and core design hit 1.000 breeding ratio anyone trying to divert either fissiles or neutrons is stuck with shutdown" HELL NO. You wouldn't do that by design. That's like saying you're going on a 200 mile trip in a 20 MPG car, so you go and get precisely 10 gallons of fuel. What if there's a diversion? What if weather is hotter needing more aircon needing more fuel? You'd be stuck by the side of the road. You would never design a system with no margins - least of all in the nuclear industry, nuclear guys and gals like their margins. You need margins to cover for uncertainties in burnup, load following, and so on and so forth. Those margins can be used or abused - as in to make weapons grade materials. Any source of neutrons can be used to make weapons grade plutonium. IFR is more neutron rich than LWRs, and has onsite fuel fab and fuel reprocessing. Arguing that this improves proliferation is silly. "That requires a higher fissile loading. Where do they get the fissiles? Chicken/egg." Where did you get the fissile to start the IFRs in the first place, smartypants? Global buildout of say 10000 GWe of IFRs would need 50000 to 100000 tons of fissile. You can't start up IFRs with the fuel that LWRs use. You need either higher enrichment, and enormous amounts of it, or reprocessed plutonium. In any case there is no need for higher fissile loading for swapping some FAs with DU fuel. You have more than enough margin in the control system, which must compensate for burnup swings, load following, startup/shutdown margins, and of course all added by generous design margins since us nuclear guys and gals are all about margins. "It's inherent in any system which can begin with 0.7% U-235. You yourself said that criticality is not an issue in gaseous UF6, and if I can design a small-diameter tubular tank set surrounded by boron or cadmium and connected by a manifold, any actual student of nuclear engineering ought to be able to do it in their sleep. " So you have allowed for modifications to be made in the enrichment plant. That is my whole argument - modifications in the IFRs (simple replacement of some FAs with DU) will get you bomb grade plutonium too. "Wrong. Pyroprocessing leaves too much in the way of fission products in the Pu fraction to allow weapons manufacture even if the isotope composition is good enough." No, wrong wrong EP. The DU fuel assemblies don't have much fission product in them, and have very good quality plutonium. The normal fuel assemblies are protected well by the FP loading, not the DU breeder assemblies. In addition, remote handling techniques today are much better than they were in the past - unlikely that a determined state that can build and operate a complicated IFR would be deterred by some FP contamination. "I'm condemning NOTHING." Yes you are buddy. Denial is not just a river in Egypt. By your fast reactor favoritism, and by spreading lack of perspective, black and white, and misconstrued arguments that constitute little more than fast reactor groupthink, you are promoting coal. IFRs aren't ready for mass deployment just yet. We need a stopgap first to deal with immediate pollution and global warming problems. I like fast reactors, but I dislike many fast reactor people that make up arguments against nuclear power that their favorite technology just might have an edge on in order to sell their concept. This is highly counter productive. " I'm recognizing a truth about nuclear proliferation" No, you are recognizing a fabrication and helping to continue that fabrication. Go read Cohen's book, its a great read. www.phyast.pitt.edu/~blc/book/chapter13.html"Are you even bothering to read what I write? That's what I've been saying: uranium enrichment (which is REQUIRED for the LWR fuel cycle) is sufficient for a nuclear weapons program." I have bothered to read what you wrote, unlike you who are only interested in your fast reactor groupthink and continual misconstrued arguments against nuclear power expansion. How are you going to get your wonderfuel fissile startup for your fast reactors? They need a heck of a lot of fuel. What, pixie dust for startup? You need enriched fuel to start lots of IFRs quickly enough to address pollution and global warming. And not the puny 5% enriched stuff that LWRs use, you need more spicy stuff. Stuff that is closer to the stuff that is in a nuclear weapon. "FBRs and epithermal reactors using a fuel cycle such as pyroprocessing which produces a recycled fuel product which is both isotopically unusable for weapons and too contaminated with "hot" fission products to fabricate into a viable bomb even if it was." More fast reactor groupthink. For probably the sixth time, simple modifications (DU FAs) will allow very good quality Pu to be produced from dedicated assemblies, and all the equipment is needed onsite. Fuel fab, fuel reprocessing, everything is there. Not much FPs in a FA that doesn't fission much buddy. Plenty of reactivity reserve in the control system to override any lost reactivity - nuclear engineers fortunately don't follow your advice on reactor design. They design for margins. Those margins can be abused readily. Any source of neutrons can be used to make weapons grade material. Moving to a more neutron rich fuel cycle, with a reprocessing plant at every reactor site, isn't helping. What would help if we all realized that power reactors aren't much of a proliferation threat to begin with; that this is an argument posited by those who would prefer to stop nuclear development, not so much by those genuinly concerned about the safety of the world.
|
|
|
Post by Roger Clifton on Jun 9, 2019 12:43:44 GMT 9.5
Where did you get the fissile to start the IFRs in the first place...? Global buildout of say 10,000 GWe of IFRs would need 50,000 to 100,000 tons of fissile. You can't start up IFRs with the fuel that LWRs use. You need either higher enrichment, and enormous amounts of it, or reprocessed plutonium. A possibly useful calculation is from "Nuclear Energy and the Environment: Environmental Sciences and Applications", ed EE El-Hinnawi, page 12 – Each gigawatt-annum of generation by an LWR produces 260 kg of plutonium byproduct. Thus 30 gigawatt-annum of LWR generation produces the 7500 kg required to provide 3000 kg of start-up fuel for a LMFBR of one gigawatt capacity and 4500 kg for replacement loadings before plutonium from its own used fuel gets recycled.
That seems to assume that every working fast core has at the same time, a similar quantity of fast fuel being fabricated and hot used fuel under water that must age before it can be recycled. That quantity would be reduced by shortening the ageing time, but I suspect that commercial imperatives would more expediently substitute 20% LEU and not wait at all.
|
|
|
Post by Roger Clifton on Jun 9, 2019 16:52:19 GMT 9.5
" Plentiful Energy" says it more simply: "almost 10 tons per gigawatt for 300 MW versus about five tons for a 1 GW reactor." That is more or less as CyrilR said it above. So a fleet of SMRs needs twice as much start-up fuel as a fleet of big fellows.
|
|
|
Post by cyrilr on Jun 10, 2019 3:48:11 GMT 9.5
“I suspect that commercial imperatives would more expediently substitute 20% LEU and not wait at all.
Yup. Grist on my mill: demonizing enrichment as proliferation hazard would be doubly damaging to a fast transition to IFRs. Both from them needing higher enrichment and needing much more of it to start than LWRs.
While demonizing LWRs as proliferation threat (on account of enrichment) is also unwise: every GW of LWR we build generates enough RgPu to start 2 or 3 GW of IFRs later.
Thus there is transitional synergy between LWR and FBR.
Framing proliferation as a nuclear power technology issue (e.g. by arguing for fast reactors on these grounds) is a loser proposition. Proliferation is a political and administrative problem, so this is where solutions must be found.
|
|
|
Post by Roger Clifton on Jun 11, 2019 18:34:45 GMT 9.5
In the event of a massive rollout of nuclear generation, there will be a bottleneck in providing the start-up fuel, in both cases. To provide the enriched uranium and/or reprocessed plutonium, there must be an expansion of uranium enrichment plants and/or an expansion in fuel reprocessing plants. Unlike the past, these future plants will be completely civil in their motivation. We will need our politicians to be persuading the global public that it will remain civil because we need nuclear electricity urgently and deep into the indefinite future.
|
|
|
Post by engineerpoet on Jun 11, 2019 23:08:52 GMT 9.5
Actually, the current pullback in nuclear generation and the surplus of uranium mining capacity would allow a large expansion of fast or epithermal reactors starting now. Harder-spectrum reactors require more fissiles per MW than thermal, but so long as they have a unity or greater breeding ratio they only require it once; there is no on-going demand to satisfy. A mere 5000 tons/year of natural uranium would suffice to start ~43 equivalents of Fermi I, or some 3.4 GW(e) of new capacity, every year. That capacity would be cumulative; 34 GW after 10 years.
The US inventory of used LWR fuel is on the order of 80,000 tons now. At approximately 0.8% total Pu and 1% U-235, that's ~640 tons Pu and 800 tons U-235. The 640 tons total Pu would suffice to start about 200 S-PRISMs at 3200 kg Pu apiece, and the U-235 would make starting charges for about 1300 Fermi I equivalents at 0.2% tails. The remnant DU would provide blanket material for all of them for several times the plant lifetimes, and that's all they'd ever need.
200 S-PRISMs at 400 MW(e) per reactor is 80 GW(e). 1300 Fermi I's at 200 MW(th) (80 MW(e)) per reactor is 104 GW(e). That's a potential of roughly 184 GW(e) without mining anything except the USA's used LWR fuel inventory.
The USA currently has no fuel reprocessing capability to speak of, so that would all have to be built from scratch. However, there might be political support for getting rid of the fuel casks at decommissioned plants, and then emptying the fuel pools at existing plants.
|
|
|
Post by Roger Clifton on Jun 12, 2019 9:56:37 GMT 9.5
To enable a rapid expansion to replace fossil carbon worldwide, there is plenty of mineable uranium to supply the world with slow or fast neutron fuel, but there is a bottleneck in the capacity of the enrichment plants. I guess full scale laser-excitation would allow a rapid scale up. Enrichment can only get cheaper. Perhaps recycling can become cheaper too. However there is ultimately a limit to the amount of used fuel lying (and accumulating) around the world. EP estimates 80,000 tonnes in the USA and no doubt there's a similar amount elsewhere in the world. If it contains 0.8% RgPu, that is potentially 640 tonnes RgPu. Even at 5 tonnes per gigawatt capacity, to be refuelled with LEU, that still only starts up 128 GW of fast generation. Of course fast reactors could start up with enriched uranium too. The Toshiba 4S is fuelled with 20% LEU. Chinese plans, for 200 GW of slow reactors by 2050, and 1400 GW fast and slow by 2100, appear to be calculated on the assumption that plutonium is generated firstly by the slow reactors, then by fast breeding later in the century. However they too, might end up starting up and refuelling with LEU. There is no doubt that the majority of observers would prefer all used fuel to be recycled, at least its transuranic content. However that is a sentiment, whereas the selection of recycled transuranics or enriched fresh uranium is likely to remain commercial.
|
|
|
Post by engineerpoet on Jun 12, 2019 23:34:24 GMT 9.5
I guess full scale laser-excitation would allow a rapid scale up. Enrichment can only get cheaper. Laser enrichment produces almost isotopically-pure U-235. That is the proliferation nightmare scenario. There's also the ~1.67%/year rate of increase in reactors such as S-PRISM ( doubling time just under 42 years with the baseline core (Table I, p. 7)). Exactly as Fermi I did. My enrichment spreadsheet says that to produce 1891 kg of 25.6% enriched U, just under 94 tons of NU is required at 0.2% tails. At a world uranium production rate of 65,000 tpy, about 690 Fermi-class units could be started annually. At 80 MW(e) apiece that's 55,200 MW(e) per year. At such a rate of increase, the entire US average electric generation could be replaced in just over 8 years. Then there's the rate of increase from breeding, which would add another ~900 MW(e) the first year, ~1.8 GW the second year, etc. That's a pessimistic figure because Fermi I was small (200 MW(t)) and had a high fissile-to-output ratio. I don't have numbers for an equivalent enriched uranium load for an S-PRISM or larger reactor, only Pu (I'm using the Dubberly differential analysis of baseline vs. high-burnup cores, linked above).
|
|
|
Post by engineerpoet on Jun 14, 2019 3:44:45 GMT 9.5
<sigh> TWICE I have been many hours into the researching and writing of this comment, and TWICE Windows 10 has taken it upon itself to reboot without authorization and destroy all of my work in progress without giving me the opportunity to save it. Windows 10 is malware. Microsoft is evil. God damn Bill Gates (and Michael Dell, whose company has both Windows 7 and a Linux distribution for this machine BUT WON'T LET ME INSTALL EITHER)! "But how many of them keep doing it on a schedule you can mark off on a calendar before the fact?" Doesn't matter. I can tell you a hundred different ways to hide this up, from sabotaging some of the plant's systems, forcing to run on part load (even 99% load would allow enough reactivity savings to make weapons) just one example. IIUC, control rods reduce power by reducing the neutron economy further. If you were going to do the same with extra neutron absorbing blanket material, you'd also have to find somewhere to squeeze it into the core. Doing that means changing the reactor structure in the core region. You think this is going to be easy? Except there's always someone by the side of the road with a gallon can of fuel... and a whole lot of questions about why you need it when you shouldn't if you were driving right. Excuse me, that's "4 litre can of petrol" to you. So you provide margins from outside the power plant site, where they are nigh-impossible to use for proliferation. Apparently I am not getting through here. 1. I am not talking about IFR. I am talking an epithermal core operated as a converter for the mass-market version. This might be feasible to do using S-PRISM as the basis. 2. Fuel reprocessing and fabrication does not have to be on-site. While that is attractive for a number of reasons, factors like not being able to trust the parties involved can outweigh them. If they aren't trustworthy, you don't let them have anything you can't trust them with. A licensed fuel production facility contracted by the vendor, most likely. The fuel might even contain a fair slug of reprocessed Pu from an actual FBR, with the FBR taking enriched U in exchange. The purpose of the reprocessed Pu is to make the fuel require handling behind shielding from the beginning. So long as the stuff is fairly radioactive, isotopically unfit for weapons, and otherwise far more trouble than it's worth to try to make it go boom, we can consider it "safe enough". 10 TW(e) is 87,660 TWH/yr, about 300 quads of electricity per year. Total world energy consumption in 2015 was just 159,000 TWh from all sources, roughly 540 quads. You might want to scale your expectations down a bit. Taking the US as a test case, it consumes roughly 100 quads per year from all sources, roughly 3.3 TW(th). At 0.32 t actinides per GW(t)-yr, it would require just over 1000 tons of makeup U per year to fully supply all energy requirements. The reference S-PRISM core has 21.29% total Pu out of 15002 kg total TRU (2458.8 kg fissile / 0.1639 fraction fissile). The US inventory of used LWR fuel would supply fuel for about 200 S-PRISM reactors at 1000 MW(t) each and 3200 kg total Pu. Those 200 units would begin breeding an excess of almost 70 kg fissiles each per year, total just under 14 tons. That's something, but not fast enough to decarbonize the economy in the time available. I don't have figures for the reactivity/neutron production equivalency of U-235 to fissile Pu in a fast spectrum. Assuming 1:1, making 15 tU at 16.39% enrichment and 0.2% tails requires 475 tNU and about 550,000 SWU. Starting 100 units per year to have another 3000 operating in 30 years would require 47,500 tNU/year and enrichment capacity of 55 million SWU. Reprocessing would have to scale up along with the operating reactor fleet. Total uranium required would be less than 1.5 million tons. Actual enrichment demand would be shrinking rapidly by the end, as about half the 3.3%/year ramp in capacity would be fueled by the 1.67%/year excess fissile production of the operating fleet. US uranium reserves are estimated to be only 138,200 tons, but that probably reflects a lack of incentive to prospect for it. Barry Brook states an EPR consumes 25.3 tLEU/yr running at 1.65 GW(e). 1.65 GW(e)/0.38 = 4.34 GW(t), so about 760 EPRs would be required to produce 3.3 TW(th). Fueling 760 of them at 25.3 tLEU/yr, 5% enrichment and 0.2% tails would require 238 tpy NU each, roughly 180,000 tpy NU total at full capacity. Over a 30-year ramp up, the LWR approach would require almost twice as much total uranium as the FBR/epithermal converter approach. Unless there is a LOT more uranium out there than known reserves, fully nuclearizing the world with thermal-spectrum reactors using LEU is a non-starter. The resource is not sufficient to supply the requirements. Serious question: why can't you create those margins using DU absorber rods, and reprocess them every time fuel is changed out? Who says you actually need margins for more than startup/shutdown and reactivity swings? Can't you handle those issues some other way? Bonus: if you're doing it with DU, neutron economy improves. How do you get sufficient reactivity to even start up if you do that, though? You claim to love margins, but where do the margins go in that scenario? There's also the issue of reprocessing capacity. The S-PRISM achieves something like 10% atomic burnup. Burning 0.32 tpyHM requires reprocessing 3.2 tpy fuel, just under 9 kg/day. But producing bomb-grade material requires pulling the blanket at something like 0.5% conversion. If there's 1% margin in the fuel cycle, you've got a whole 90 grams of excess capacity from which you can extract maybe 0.45 gram/day of bomb-grade Pu (plus included FPs). That's going to take decades to get just 1 critical mass. If reprocessing is done off-site, somebody is going to notice the anomalous blanket material and know something is up. Per Dubberley, the S-PRISM inner blanket has a peak burnup of 29.3 MW-d/kgHM, and the radial blanket and fuel are far higher. I wouldn't call that "low", and it's several times the maximum possible burnup of the MAGNOX fuel used for the (failed) bomb test of reactor-grade Pu at the Nevada Test Site in the 60's. Unless you're swapping out radial blanket elements pretty rapidly you're not going to get bomb-grade Pu in them. Axial blanket, forget it. So what is spreading LWR fantasies in denial of things like uranium resource limits? I've got nothing against a bunch of e.g. AP1000s as an immediate effort. However, even if we got the schedule down to 5 years we could have a preliminary S-PRISM fleet in operation before there were more than a couple dozen of them. You also run up against uranium resource limits by the time you hit the TW(th) level. The USA ran a highly successful LMFBR from 1964 to 1994. The S-PRISM is a direct descendant of that reactor. If I were energy czar, I'd let a contract to build 3 of them with the first 2 for practice and the 3rd being the one used for estimates of fleet cost. I like molten-salt reactors but I have to acknowledge that they are far less market-ready than LMFBRs. We need a good 10 years of experience at the 100 MW(t) level before we can think about going big with MSRs, so that would be 2035 even assuming a crash effort starting today. I'm telling you about resource limits of uranium, and you link me to Cohen's chapter on freaking plutonium?! Are you for real? That's doubly ironic because I regularly cite Chapter 9 when Greenies say that nuclear is "too expensive". It's the government and the lawyers that cost a lot. Whereas you deny that uranium enrichment has anything at all to do with weapons proliferation. Yup. I'd need something between 17% for S-PRISMs and 25.6% for Fermi-I equivalents. I could start maybe 180 GW(e)/450 GW(t) by reprocessing and re-enriching the US inventory of used LWR fuel, and an unknown number by re-enriching some of the higher assay DU tails. And at that point, at roughly 13.5% of total US primary energy consumption, those resources would be gone. I'd have 2 options left: 1. Rely on growth from the surplus fissiles produced in the FBR fleet, at roughly 1.67% increase per year. That would add about 7.5 GW(t)/year at first, far too slow to decarbonize the remaining ~2800 GW of primary energy consumption in time to matter. 2. Use enriched U as starting charges for new LMFBRs. Yes, that would require a lot more in each initial fuel charge... but it is only required ONCE, and then the new unit starts adding that 1.67% per year without adding any more. And for probably the fifth time, the IFR all-at-site fuel cycle is not a design requirement. Neither is an over-unity breeding ratio. GEH says it's ready to start building S-PRISMs for anyone able to put up the money. That's the sort of offer we need to jump on. So don't reprocess at every reactor site. There was no such thing at Fermi I, and no such plan for S-PRISM. Oh, FFS. <rolls eyes> For probably the tenth time, it isn't the REACTORS that are the proliferation threat. It is THEIR FUEL SUPPLY CHAIN. This supply chain is tightly controlled world-wide, for good reason. This presents 2 major problems: 1. Scaling it up radically means more difficulty keeping tabs on it and more likelihood of diversion by bad actors. 2. Limits of the known resource suggest that it may be IMPOSSIBLE to increase things to the required level using uranium-fuel thermal-spectrum reactors. (Thorium? Sounds good, but we need at least 15 years to prove new fuels before we can expect to rely on them and we need to be well on our way by 2035.) There are a bunch of new-tech nuclear designs out there. NuScale. Moltex. ThorCon (which now has a new name). S-PRISM. NuScale and S-PRISM are the most advanced, and S-PRISM is the only one which can get us out of the uranium resource bind. Like I said, it's time to build at least 3 (plus the fuel handling stuff) and see what we can learn or re-learn.
|
|
|
Post by Roger Clifton on Jun 14, 2019 7:17:40 GMT 9.5
Limits of the known resource suggest that it may be IMPOSSIBLE to increase things to the required level [ ie for nuclear to displace most fossil generation ] The concept of "limits of the known resource" is more familiar to us as a religious tenet for the renewables people. They believe that all mineral resources have been identified, measured and are currently being mined. With that as an unquestioned premise, it follows that the industrial world is running out of all mineral resources on the same time scale that mining companies measure their reserves. Geologists say otherwise. Uranium is as common as the dirt under our feet. Its average in the continental crust is two grams per tonne -- inexhaustible. Being quite soluble and chemically active, it forms more concentrated deposits in many places. It is so easy to mine that its current price is less than $100 per kilogram. Fully fissioned, a kilogram of uranium would provide a thousand kilowatt-years of electricity, so the scarcity cost of the fuel in nuclear electricity is negligible.
|
|
|
Post by engineerpoet on Jun 14, 2019 12:57:35 GMT 9.5
Uranium is as common as the dirt under our feet. Heck, it's present in substantial amounts in the very Marcellus shale from which Pennsylvania is extracting the current glut of natural gas. That's where all the radon in it is coming from. That doesn't mean we have any easy way to get it out. Maybe we do, but so far I haven't seen anything that mentions how.
|
|
|
Post by David B. Benson on Jun 14, 2019 19:53:15 GMT 9.5
engineerpoet --- Thorcon Power continues to use that name, via websites.
|
|
|
Post by engineerpoet on Jun 15, 2019 0:41:31 GMT 9.5
Okay, I did some digging and came up with this paper on radon in Marcellus shale gas: www.ncbi.nlm.nih.gov/pubmed/26882276Weighted average concentration is 1,983 Bq/m³. I'm at a bit of a loss of how to go from this back to a radon generation rate in the source shale and from that to an estimate of the uranium concentration, but it's a start.
|
|
|
Post by engineerpoet on Jun 15, 2019 0:44:23 GMT 9.5
|
|
|
Post by cyrilr on Jun 15, 2019 3:59:54 GMT 9.5
Oh boy. Nonsense galore from someone who calls himself an Engineer (but a poet too, so that explains things more).
Yup. Easy peasy, lemon squeezy. Plenty of margin in control rod design. Us nuclear guys love our margins. Can make several Pu bombs from those margins if you're malignant.
Did you know that in the USA you need a 50 mk shutdown margin? Wonderful, that can be used nicely to make bomb grade materials.
The simple fact you're avoiding is that load following requires some control rod worth, and this worth can be used either for load following OR for making bombs. Margin is needed for operational and licensing reasons. 1 mk margin will make you a couple of bombs over the facility lifetime.
Yup. And no IAEA to ask any questions no matter how much gas you fill. Transport always gets a free pass. Nuclear power, not so much. You have to justify every freaking gram of fissile buddy. And you are avoiding the argument: going without margins is a no-no in the nuclear industry. Forget it. Ain't gonna happen.
no can do buddy, margins needed for operational reasons and for licensing and design... needs to be in the core plus on site added fuel. Forget about your fantasies. You obviously don't work in the nuclear industry.
More elitist nonsense from a first world technocrat. 1 kWe per person for 10 billion people, clearly that is too much for you to propose. I guess you want to keep the world's poor poor so you can continue your demogogue sprea and continue to play the elitist rich kid on the block? Screw you buddy. World's going to 10 TWe in no time regardless of what you conject upon.
Maybe it is because you prefer conjecture, group-think and demogogy over simple facts presented to you in earnest?
Or is it because you chose to focus on peripheral details rather than the main argument at hand?
The fuel would have to come either from higher enriched fuel or from reprocessing RG Pu. Simple fact that you are eluding in order to feed your fast reactor group think of "we don't have any proliferation threat, no, despite our neutron rich fuel chain proposing onsite fuel fab and reprocessing that is easy for diversion". Oh yeah. Very convincing.
Sure you could, and get the nice DU diverter Pu production elements in the design in the first place. Non proliferationists will love you.
DU rods have low worth so would be of limited control value. The same reason why DU rods would be a proliferation threat. You need 7 kg to make a bomb. You got 4000-8000 kg/GWe fissile inventory. Go figure.
I'm not talking standard fuel cycle, I'm talking would-be-proliferationists that simply swap out specific fuel elements for DU elements. You can easily get weapons grade Pu that way even in normal burn cycle. Not "inner blanket", it would be an outer "reflector" of DU assemblies. Possibly not even a complete circular reflector... not much Pu is needed for a bomb and a fast reactor has 1000 bombs worth of fissile in it.
You mean the resource limits you just incidentally and conveniently made up? Congrats you're a fast reactor group-thinker demogogue. No such thing as running out of uranium. Go read my nuclear FAQ or go read Deffeyes paper. We won't run out this millenium. But please go on and fabricate more arguments in favor of fast reactors, it is most amusing to me to see my world view being confirmed even by people who should know better.
You are only just bringing up this point, which by the way is wrong - we can increase LWRs 10x and not worry about uranium resources this millenium, go read my FAQ - but it is most amusing for me to see you bring up heretofore unbrought-up points in order for your argument to last the day - sadly for you it is another fabrication. You might do good working for the Sierra Club or Greenpeace, they would love you.
So you finally admit to being screwed - fast reactors needing both more fuel and higher fissil concentration to begin with - to get going.
I don't feel you are interested in an honest debate or in even remotely comprehending the simple arguments I laid forward. Please prove me wrong.
|
|
|
Post by Roger Clifton on Jun 15, 2019 13:41:09 GMT 9.5
Extraction from black shales (such as the Marcellus shale, 10-80 ppm U) is quite feasible especially if localities are chosen for a uranium content of ~300 ppm or more. Uranium is routinely extracted from ore as low as 500 - 1000 ppm. Considering that minesite metallurgy routinely extracts gold, a noble metal from ores as low as 2 ppm, it only behoves the chemists to juggle the special chemical behaviours of uranium to extract the pure product.
Uranium is more soluble in oxidised and acid solutions, so its concentration is found high in reduced ground such as black shales and coal) and alkaline ground (such as near-surface calcrete). Its electron sphere is particularly large, so it is commonly found in nature as contaminants in large anion cages like orthosilicate (e.g. zircon) and orthophosphate (e.g. rock phosphate). For the same reason, it attaches to surfaces and colloids, deposits with ironstones, and travels down rivers on the humic content. In other words, uranium is everywhere.
Mined uranium is powdered and leached out (with its daughters) with a strong acid and oxidant. A suspension of kerosene droplets is applied, carrying a special detergent with wide phosphate jaws, which targets the uranium, trapping it onto the droplet. It is then stripped out and precipitated with ammonia (the classic process producing yellowcake) or sodium hydroxide (which precipitates out the white oxide).
In the mining method, the residue contains the daughters so rehabilitation of the minesite requires special environmental vigilance to ensure that they stay put forever. When the ore is leached while still underground, the daughters are left in place, but inevitably some of the solvent is left there too, a booby-trap for future explorers. When uranium is extracted as a byproduct from other process streams (such as superphosphate), the daughters may be or may not be encapsulated and buried, but may also left in the mainstream (fertiliser that is eventually applied to the fields). When uranium is extracted from seawater (by flowing across an activated surface), the daughters have already been dumped in the wide ocean by Nature herself. A process intended to provide energy security for Japan, seawater extraction is more expensive than current mining methods. Currently, that is. We ain't gonna run out.
|
|
|
Post by Roger Clifton on Jun 15, 2019 15:07:32 GMT 9.5
1 kWe per person for 10 billion people... World's going to 10 TWe in no time According to WNN, the Chinese plans for the nuclear future are to provide 1400 GW of fast and slow generation by 2100. Their population may or may not then be in the vicinity of 1400 million people. That is, 1 kW per person, as currently in developed countries. I want the world to do better than that, to decarbonise all energy consumption. Providing power for heating, cooking, EV's, synfuels, smelting and cement is likely to triple that requirement to 3 kW each. In that scenario, 10 billion people would require 30 billion kilowatts, 30 TW. Current calls for action refer bravely to the year 2050 as the target for full decarbonisation. Presumably shortfalls and recalcitrance will still take us past the year 2050, but with clarification of the sacrifices and degree of mobilisation required. Achieving 30 TW by 2050 would require 937 GW of capacity installed per year. Achieving it by 2100 would require 375 GW/a. It is remotely possible that between now and then a deep voice will come out of the clouds, saying, "Well dear frogs, how do you like the nice warm water? Are you listening to me? Hey, wake up!" Then, after a particularly nasty climatic disaster, another deep voice in the clouds will say, "Well how did you like that one? Welcome to the future. This is it. Do you want it to get worse?" On the other hand, by that time the complacent majority, the bad guys, you and I will have collected our fat superannuation from oil and gas etc and begun to fade out. The solutions are then left as a problem For The Kiddies. Senile by then, we might mumble our salute to the future, "FTK!"
|
|
|
Post by engineerpoet on Jun 16, 2019 1:29:21 GMT 9.5
Extraction from black shales (such as the Marcellus shale, 10-80 ppm U) is quite feasible especially if localities are chosen for a uranium content of ~300 ppm or more. One paper I found mentioned that fracking often uses HCl to dissolve carbonates and open up pores, and also used HCl to extract the U from the samples for the assay itself. It might be possible to use the fracking process itself as a method of mining uranium. It doesn't matter if the yield is low so long as the added cost is small. My reply to CyrilR is on the way.
|
|
|
Post by engineerpoet on Jun 17, 2019 6:14:55 GMT 9.5
Oh boy. Nonsense galore from someone who calls himself an Engineer (but a poet too, so that explains things more). Descent into name-calling right out of the gate. Classy. Sets the expectations for what comes later. You're not interested in ANYONE comprehending what you say when you use jargon like "mk" without reference to what it means in your specific context. Do you have ANY idea how many things are abbreviated "mk"? Try shedding light instead of radiating heat. Light is useful, heat is a waste product. And they DON'T need isotope separation in operation. They can merrily consume stuff that's useless for making bombs, and make more of it. With a hard enough spectrum they can even consume americium and curium, turning a waste disposal problem into fuel. Okay, enlighten all of us rubes out here. HOW? You're going to have to explain the term "mk" before that's going to be any kind of illumination to all us rubes. Is that what people have been calling "cents" since roughly the Chicago squash-court pile? First, two bombs-worth over 60-80 years isn't exactly the sort of breakout that worries me. Second, that's also subject to constraints on isotopic composition; N. Korea proved that you can produce enough fissiles and still not have the makings of even a single usable weapon. I was making an analogy. Apparently you missed it. However, you've just contradicted yourself. You were specifically making a case that it was EASY to create and divert bombs-worth of fissiles without anyone being able to account for them or detect the act. Now you're saying "every freaking gram" is accounted for. Well, which is it? This is something that probably calls for a post all by itself, and I encourage you to write it. You said that those same margins allow the undetectable creation and diversion of weapons-grade fissiles, which would use the margins in an un-approved and I suspect not very controllable manner. Isn't that "doing without them" at least functionally? Quite right, I don't and never have. This is a hobby with me, and I've never pretended otherwise. Now, how about explaining those ins and outs that you are privy to, but the rest of us haven't learned about yet? I don't think so. A large part of that projected 10 billion are people who had tens of thousands of years and never invented the wheel or sailing boats. Where they try to take charge of their own development, they rapidly descend back to the stone age. Most of those countries aren't even self-sufficient in food! They aren't going anywhere without lots of help, and the rest of the world is tired of dealing with them. If I hadn't gotten some weird messages that prompted me to shut down everything except Notepad and run two full virus scans (second one only 20% complete so far), I'd go back to Brooks' piece on the uranium requirements of Gen 3+ LWRs and see just how well they'd scale up to your claimed 10 TW(e) level. Because if it isn't done with nuclear, the coal and natural gas required is gonna wipe us out pronto. You mean, like the simple facts about the EPR, S-PRISM and Fermi I that I linked for everyone, unlike the name-dropping of Deffeyes (not even a quote!) that you count as a rhetorical killshot? You mean, like the selling point of pyroprocessing is that its products are useless for weapons? How many times do I have to repeat "I am not talking about IFR" and "Fuel reprocessing and fabrication does not have to be on-site"? Do you think you could substitute raw DU rods for the irradiated ones in the shipment to the reprocessor and nobody would notice? It sounds like just what you need to manage the relatively small reactivity swing of the baseline S-PRISM core. The reactivity increases from BOC to EOC, so you could insert DU rods incrementally to keep the reactivity where you want it. At EOC the rods get reprocessed along with the blankets and replaced with fresh ones. Needless to say, this is NOT for the mass-market epithermal model. Definitely not for the mass market. A possibility for the established industrial members of the nuclear club. Not following you here. 1. Swapping out fuel (fissiles) for DU means a substantial drop in reactivity. How much of this can you do before you can't get to rated power? How much before you can't even start the reactor? 2. If the fuel must be sent out for reprocessing, can you do this without sending up great big red flags that say "I'M DIVERTING FISSILES"? Fuel elements with anomalous burnup would do that. The point, my flame-ish friend, is that the fissiles are mixed up with a whole bunch of stuff that's not fissile, that's got heaps of spontaneous fission neutrons coming out of it, radiates gammas that chemically alter things like compression explosives and maybe make them unstable, and lots of it cannot be separated by chemical means because it's unwanted isotopes of the essential element involved. Taken straight from here: en.wikipedia.org/wiki/List_of_countries_by_uranium_reservesUS reserves are quoted as 138,200 tons as of 2015. By my calculations, that's enough to make just 1470 Fermi-I cores, only 294 GW(t) worth. Of course, if you run it through LWRs first that drops to something around 1/3 as many and it's a long, slow slog to breed back up to where you started. Except I'm not a member of any "group"; I came to my conclusions quite independently. I haven't even mentioned anything about these thoughts to Rod Adams or Charles Barton yet. I'm just thinking about long-term sustainability. I've read some of Deffeyes' stuff before but name-dropping is not actually citing him as a source. Try linking, it works better. There are words for someone who takes a claim with a reference to an independent source and calls it "fabricated". They include "sad", "slipshod" and "dishonest". Your concern about fissiles for fast reactors seems like projection. A LWR takes LEU at perhaps 4.5% fissiles and yields perhaps 1.5% fissiles in its product. The baseline S-PRISM core increases its fissile inventory by about 140 kg per 24-month cycle. Shouldn't we be doing that, if not instead, at least in addition to what we're doing now? That's one of the things I lost due to my malware (Windows 10) infection and unauthorized reboots. Adobe Reader does not keep a list of external files to restore. Plus, you never did provide anything like a change log to keep your reviewers from having to go back over sections that had no new material. I had to start at the beginning every time you put out a new version, and had to give up due to lack of time. As if! I got banned from Cleantechnica for making ONE comment there. And if you mention reprocessing nuclear fuel to get rid of the "waste problem", they practically need fainting couches. (Note, at this point I'm writing in stream-of-consciousness mode while waiting for a virus scan to finish so I can open up other apps again. Just getting thoughts down while killing time.) While you're ignoring my references and talking trash, I'm pondering how to de-fossilize the entire US economy. Part of that is going to be the political end. Anti-nuke propaganda has made lots of people worry about used LWR fuel at reactor sites, both active and closed. Reprocessing all of that fuel is one good way of getting rid of it. Building a new system which consumes the "waste" from the old one, has minimal inventories of its own and puts the country on a path to zero emissions and total energy independence just might be palatable to the public. Attractive, even. So far as spent LWR fuel is concerned, I would start with the better-aged stuff in the fuel pools of active reactors. The anti-nukes have been whipping up hysteria about spent fuel catching fire in a disaster because the pool runs out of water and the fuel is too hot and too densely packed. Fine, start taking fuel out of there for reprocessing and reduce the density back below the original spec. That also eliminates the issue of where to move it, either temporarily or permanently. It doesn't sit around any more, it gets recycled. I still think that the ~80,000 tons of used LWR fuel is, if not key, the proper starting point for any sustainable solution to this problem and fast-spectrum reactors are essential to dealing with it. First, that fuel has no real other use; it cannot fill in for other resources. Second, the contaminants like U-236, Pu-238 and Pu-240 are not problems for FBRs. Third, 640 tons of recoverable U-235 and another 640 tons or so of total Pu would start quite a few FBRs. The recoverable, re-enrichable uranium would supply starting charges for about 1320 Fermi-I equivalents (200 MW(t), 484 kg fissile U-235 each), total 264 GW(t). The plutonium would suffice to fuel roughly 200 S-PRISM units containing 15002 kg TRU @ 21.29% total Pu; that makes another 200 GW(t), total 464 GW(t) and maybe 186 GW(e) of emission-free energy.
|
|
|
Post by huon on Jun 17, 2019 6:51:23 GMT 9.5
***MODERATOR***. CyrilR and EngineerPoet-- Tardy though this moderation is, we've got to get back to civility on this thread.
Update (~3 hours later): Please read the BNC Forum Comments Policy, by Barry Brook, at the top of the Forum Help section. A excerpt:
"Civility--Clear minded criticism is welcomed, but play the ball and not the person. Rudeness will not be tolerated. This includes speculation about motives or what 'sort of person' someone is. Civility, gentle humor and staying on topic are superior debating tools."
Note: This post was begun, and the first sentence posted, before I saw your latest post, EP. I am not singling you out for blame. Indeed, most of the culpability for the "corium" in this thread I place squarely on myself: If I had intervened sooner all of this could have been avoided. I offer my apology to both of you, who are valued members of this site.
|
|