Transatomic Power's Safer Reactor Eats Nuclear Waste - Businessweek
Here's an interesting Gulch like take on power generation, make it from our existing stock of "garbage" or waste. How Galtish is that? Of course the people who have to approve it have "no way to review novel or new designs". So what did they do in the 40's and 50's?
I would like to take this opportunity to quibble with the term 'waste'. This is a good example of derisive labeling. "What are we going to do about all of this horrible nuclear waste?" people wail. It is not waste, it is byproducts; it is not something that needs to be gotten rid of but something that needs to be preserved as a resource. To whit: use in seeding the reaction in the next generation of reactors. To call it 'waste' is to fall smack dab into the agenda of people who are against nuclear power.
I would also like to mention that, back in the 60's, before the environmental movement shut down nuclear development, there was discussion as to whether, with nuclear plants as the source of electricity, it would cost more to bill for electricity than the effort would be worth. India may be the nursery of the thorium reactor but we would all like to have the benefit.
Jan
Zircaloy has more radiation damage resistance than alternative materials, but it still has its limit of useful life before a locally weak spot cracks and starts leaking radioactive fission gas byproducts into the cooling water. When such a leak is detected, the fuel assembly containing the bad fuel rod is removed, even though the other fuel rods in the fuel assembly still likely have some useful life left in their zircaloy cladding.
The used nuclear fuel of the fuel rods in the "spent" fuel assembly still contain a significant amount of actinides and heavy metals to support additional fission energy release. To more fully utilize that fuel, it could be extracted and reprocessed into a form that some of the advanced reactor concepts can use.
Pumping a molten, heavy metal fluoride salt mixture through the reactor's primary loop and going through a heat exchanger to dump heat to the secondary loop (which uses either a Rankine or Brayton cycle to drive the electric generator) introduces a couple of big problems that nobody has wanted to tackle with much R&D spending:
1) Severe chemical corrosion of all components in the reactor's primary loop (pipes and the intermediate heat exchanger) is a big life-limiting and maintenance problem. In a Gen II (current) or III (being built) reactor (light water reactors), chemical corrosion of pipes from cooling water is the biggest maintenance challenge (significant cost) over the typically 60 year life of a nuclear power plant. Using fluoride salts makes all of these material compatibility and maintenance issues an order of magnitude more problematic.
2) Pumping a radioactive, molten fluoride salt mixture through the primary loop pipes and intermediate heat exchanger degrades their material properties faster (in addition to the corrosivity problem) due to radiation damage of the metal alloys in the structural material. This significantly shortens the useful life of the structural material, requiring more frequent replacement of pipes and intermediate heat exchanger (difficult inside the reactor's radiation containment...).
There are other less challenging problems with MSRs, but the 2 above are the ones that have been show-stoppers to any significant R&D investment over the years. There are 5 other Gen IV reactor concepts receiving most R&D funding, due to being less challenging, etc... Two of those also help close the nuclear fuel cycle (minimize waste) but are breeder reactors (a proliferation concern since the plutonium can be extracted for use in weapons) -- LFRs and SFRs:
1) Lead-cooled Fast Reactors (LFRs), being advanced by Russia in their BREST family of reactors, use more conventional solid nuclear fuel elements in the reactor core that mix in used nuclear fuel. The coolant is liquid lead instead of water. Molten lead has a high heat capacity, making it thermally a great coolant. However, lead has a high melting point. If the temperature of the lead drops too much, it starts solidifying inside all the pipes of the primary cooling loop (which usually have heaters to try to avoid this...).
2) Sodium-cooled Fast Reactors (SFRs) advanced mostly by the U.S. and a few other countries are similar to LFRs but use liquid sodium as primary coolant. Sodium has a significantly lower melting point than lead. The drawback is the risk of sodium fires, if hot sodium comes in contact with air (like if a pipe springs a leak at a joint). So, the reactor core has an inert atmosphere and inspections are more frequent for pipes, etc... The U.S. DOE successfully built and ran test SFRs (EBR-I and EBR-II) from the 1950s through 1994. While successful, they were expensive projects.
I think the likely long-term future combination for U.S. nuclear power plants will use a mix of 2 types -- VHTRs (commercial) and SFRs (DOE). For closing the nuclear fuel cycle, a DOE-operated SFR (with required adjacent used fuel reprocessing plant) would be dedicated to more fully utilizing spent nuclear fuel from commercial power plants. For commercial power plants, VHTRs (Very High Temperature, gas-cooled Reactors) would operate with higher efficiency than Gen II light water reactors. VHTRs, due to the higher temperature of the helium gas exiting the reactor core, can also be used as process heat for hydrogen production and other industrial facilities, in addition to electricity generation. R&D on VHTRs is much further along than most of the other 5 Gen IV reactor designs.
Of course, the biggest roadblock to innovation is the U.S. Nuclear Regulatory Commission (NRC, which is separate from the DOE). Any new reactor design takes at least 8 years (usually far more) to obtain NRC licensing approval for commercial construction and operation. The nuclear power industry is the most heavily regulated industry in America, which results in a depressingly slow pace of innovation...
Fascinating. Thank you for detailing the issues in a most comprehensive way. As a layman I really appreciate your explanations. Question: Could materials other than metals be used to provide the piping for the fluorine in the MSR reactors, thus mitigating the corrosion problem?
This problem seems similar to a problem common to investment casting plants that use Kolene to dissolve ceramic from inside their castings. It too is a molten salt bath. The process is so corrosive that it will dissolve aluminum castings, and can therefore only be used on certain steel alloys. I don't know what material the units tanks and piping are made of, or lined with, but I know they have been used reliably for 75 years... If this problem could be overcome, would this be the best type of reactor compared to the others you have discussed?
Respectfully,
O.A.
As good as high strength stainless steels are for corrosion resistance, they still require inspection and replacement in current Gen II/III light water reactors. I should point out that corrosion rates, like most chemical reaction rates, increase significantly with increasing temperature of the reactants. Thus, for a power plant design, you want to operate at temperatures as high as practical. For designs more susceptible to corrosion reactions, you either limit the operating temperature to limit the corrosion reaction rates, or you look for alternate corrosion resistant materials/coatings, or operating conditions that promote formation of stable corrosion byproducts (oxides, or fluorides for MSRs) that create a protective layer of scale that inhibits further corrosion.
I'm not the materials or corrosion expert, so I don't know what other alternatives may exist to handle high temperature, corrosive fluorides at modestly high pressures. I would guess there are few alternatives to high-strength stainless steels currently used and would be more expensive.
Again, I wish those MIT students the best of luck. But when you develop new power conversion design concepts, the last things that usually get attention and resolution are issues of maintenance, operability, reliability, and total life cycle cost. The thermodynamic conversion cycle is easy. The reactor core physics work gets most of the early development. But the early estimates of life cycle cost always underestimate costs associated with maintenance and reliability of components and any oddities associated with operability or complications during construction.
Yes, an alloy of stainless steel would be my first choice, although I am always hearing about new plating/coatings that can be applied including diamond. I'm sure there are many factors to consider including conduction, convection, etc. If a particular alloy of stainless will do the job economically, I would design with a redundant system for emergencies as well as maintenance so components could be changed out at necessary intervals. Certainly cost is a factor to be considered... unless it is our government spending tax dollars of course. :(
Excellent exchange.
Regards,
O.A.
http://www.haynesintl.com/historypage/hi...
There have actually been a number of newer better superalloys invented since the LFTR at Oak ridge was shut down. If they could operate walk away safe back in the 60's and 70's with material science from 40+ years ago it should not be hard to blow that success away..
http://www.haynesintl.com/pdf/h2052.pdf
The corsion problems with lithium salts at high temperature are pretty much just engineering
problems.
It seems that the biggest problems they need to do a bunch of solving for involve chemically separating the reaction products from the lithium salt and the remaining actinides. Some of the *waste* products from these reactors include some really useful rare earth elements. Also, it there will be some medically useful isotopes. The rest of the actinides should probably remain in the reactor till they are consumed.
This video is worth taking a look at..
https://www.youtube.com/watch?v=8Pyq8kCe...
The video was quite informative. According to the video, it would appear that the pencil pushers and bureaucrats have stood in the way of development of safer alternatives despite many engineer's and nuclear physicist's alternatives. One must have funding to see anything accomplished and the tried and true is winning. In this field we are still using old technology... If computer and phone tech, developed in this fashion we would all be still using floppy discs and rotary phones...
Regards,
O.A.
The video looks very interesting. I am going to try and view it in its entirety tomorrow morning.
Thank you.
O.A.
Yes, Hastalloy is very tough. It is used in Wheelabrators in the foundries I support for shot blast/peening, finishing and ceramic removal. In the past, I produced several molds to produce patterns for casting of Hastelloy Wheelabrator parts (Blades/vanes, impellers). It does have mechanical properties that suggest possible application. It is very abrasion resistant, but I am unfamiliar with its salt/corrosion resistance characteristics.
Regards,
O.A.
As a nuclear engineer - although granted with only a year of experience before the industry really went dark (pardon the pun) in the early-mid 90's - I was a big fan of the Integral Fast Reactor concept, an SFR (to use robertmbeard's terms) using liquid sodium first and second loops, metal fuel (vice ceramic) and on-site reprocessing via pyroprocessing concept. LFTR is a better concept by far, with most of the corrosion questions not looking nearly as insurmountable as 60 years ago.
If MSRs, which have a lot of good features, ultimately can be developed and result in a superior combination of life cycle cost, safety, fuel utilization, etc., than it should be the preferred reactor design. I personally was excited to learn more about MSRs when first introduced to them.
There are a few reasons I specifically highlighted VHTRs and SFRs as most likely. First, they have received the most R&D funding in the U.S., and to varying extents, the most test reactor experience (especially EBR-I and EBR-II). So, they are closest to full development with better characterization of design issues, etc. for use either in commercial power plants or in a DOE-operated facility (in the case of SFRs).
Secondly, politics and bureaucracy have stifled innovation and progress in the U.S. commercial nuclear industry for years (due to varying extents to the NRC, environmental activists, DOE, politicians, etc...). If I had to bet money on any Gen IV reactor concept being built in the next 15 years, I would not given all of those impediments to innovation (especially the NRC). Keep in mind it is "illegal" in the U.S. to build and operate any nuclear reactor without mountains of licensing paperwork to/from the NRC. The DOE national labs and the NAVY nuclear program (subs, ships) do not fall under NRC's jurisdiction and are the only areas where innovation has very slowly occurred over the past 60 years. So, as wrong as it is, possible advanced reactor concepts like the MSR have to not only overcome the many technical design challenges, but also DOE R&D funding bias against the concept, NRC ignorance and licensing bias, environmentalist lawsuits, politicians abusing power, etc...
I hate to sound pessimistic about future nuclear progress in the U.S., but it is the most heavily regulated industry (commercial power) and doesn't fit the political agenda of most environmentalists (despite having virtually no carbon emissions that they focus so much on)...
Good coolant characteristics include high heat capacity, low pumping power requirements at the operational flowrate, good thermal conductivity, etc...
There are important neutronics characteristics required, as well, of the coolant. In a typical light water reactor, which requires fission neutrons to be slowed (moderated) to lower speeds and energies to fission the most U-235 fuel, the coolant needs low to zero neutron absorption cross-section and high neutron scattering cross-section (so collisions transfer energy and slow down the neutron).
In most advanced reactors that have the best nuclear fuel utilization, most operate with a fast (high energy) neutron spectrum. In that case, any coolant needs low neutron absorption and low scattering characteristics, so that the neutrons maintain high energies.
There are other design considerations tied to the coolant choice -- cost, corrosivity, flammability, safety features needed to mitigate or avoid accident scenarios, etc... Like any design choice, there are tradeoffs made.
All Gen III and IV reactor designs feature passive safety features and redundancies to avoid the safety deficiencies of past reactor failures. And any commercial power plant design is required to operate safely, reliably, and economically for usually 60 years.
Gas coolants such as helium in VHTRs do not have flammability or corrosivity concerns, but gases have inferior heat capacity to liquids for use as coolants. The biggest benefit for reactors using certain gases is that higher temperatures and operating thermal efficiencies (and electrical generation efficiency) for the overall power plant can be achieved. The associated downside is the high temperature material property limitations of materials used by parts of the system and, in some cases, the cooling challenges during emergency shutdowns.
Fluoride salts in MSRs are highly corrosive, which is beneficial for dissolving used nuclear fuel into the mixture but more challenging for long-term corrosion of structural materials in the rest of the system.
Sodium has superior (near zero) neutron scattering and absorption cross-sections for use in a fast energy spectrum, breeder reactor. This makes achieving criticality easier and contributes to better nuclear fuel utilization. It is also a superior coolant. The annoying downsides for sodium are the safety risk of a sodium/air or sodium/water exothermic reaction (fire) in the event of a leak into a non-inert atmosphere. Thus, the design and operation of an SFR is more complicated in order to avoid or mitigate that issue. The melting point (206-208 F) is also just above typical ambient temperatures, which creates a risk of solidification of sodium in pipes for a system that is shut down long enough to cool down. That design challenge is addressed with pipe heaters, etc...
In the normal evolution of product designs in other industries, every successive generation of the product addresses or improves on the problems of the previous one. So, through steady learning and innovation, better designs are matured. The DOE has funded a significant amount of basic nuclear research and demonstrated concept viability for many reactor design concepts, as a result of testing in the 50s, 60s, 70s, and to a lesser extent afterwards. But due to various factors (like NRC regulations) mentioned previously, the pace of U.S. nuclear power plant innovation has been depressingly slow, since the normal evolutionary design maturation process occurs in fits and starts, or not at all...
It seems to me that if we wish to remain competitive in wold power generation it's time we turn loose the productivity and imagination of Americas industry.
recommended by robertmbeard, on this post, crossed
with a thorium reactor using set-aside (Not Waste)
fuel rod pellets, to prove it in. then, strong-arm the
NRC with evidence that it works. rotating blackouts
produced by BHO's curtailment of the coal-power
industry should get the public on our side.
if it were mine to design, I would have multiple
fuel flow paths, multiple coolant flow paths, and
multiple cooling tower flow paths -- to allow long
runs during alternate-pipe maintenance.
too expensive? how expensive are shutdowns?
just thinking ! -- j
project UF6 plant (6.5 years there), where the
gas solidified when shut down. now, this does not
result in a wall-to-wall solid, I admit, but preparation
for this is not impossible. 208F is the melting point
of Na,, so you engineer in heated chambers for
the piping --- and crank up robots to do the
maintenance! cubic $$, yes. I wouldn't think it
smart to think of this for a sub.
maybe I'm missing something. -- j
p.s. Yes, the primary obstacles are politics, the
media, and the lawyers.
But the real goal--and let's not kid ourselves--is fusion.
As for fusion, the old joke joke about fusion always being 50 years off is quite descriptive. At least with tokamak type designs you could envision a heat transfer mechanism to deliver power out of it. The only fusion mechanism enjoying success at the moment is the laser compression, and how do you make a power plant in that configuration???
This is a more interesting approach that vitrifying nuclear waste (encasing it in glass), which is the traditional way.