Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Transatomic - Molten Salt Reactor (from Question about LFTR?)

  1. Feb 19, 2015 #1

    mheslep

    User Avatar
    Gold Member

    Transatomic is recent nuclear startup by a couple of MIT nuclear engineers which has so far gathered a few million dollars in funding. Their particular design approach uses molten salt for a fuel as does the LFTR proposal, but eschews thorium for a uranium only design. The Transatomic http://transatomicpower.com/white_papers/TAP_White_Paper.pdf [Broken] thoughtfully anticipates questions from thorium advocates in a Chapter titled "Why Not Thorium First?". Excerpt here:

     
    Last edited by a moderator: May 7, 2017
  2. jcsd
  3. Apr 9, 2015 #2

    Astronuc

    User Avatar
    Staff Emeritus
    Science Advisor

    I created a new thread on this topic since it represents an innovation of the MSR concept.

    http://transatomicpower.com/company.php [Broken]
    http://transatomicpower.com/press.php [Broken]

    Presentation at CNA 2015 - https://cna.ca/wp-content/uploads/2015/03/Leslie-Dewan-Transatomic-CNA-Feb-2015.pdf (link is subject to change or disappearance)

    https://www.haynesintl.com/pdf/h2052.pdf [Broken]
     
    Last edited by a moderator: May 7, 2017
  4. Apr 13, 2015 #3
    Its really neat that the design is progressing. I really like that they have dropped thorium for the first generation design. While it has potential long term, it makes sense to keep the first generation of MSR as simple burner reactors especially if they are targeting spent fuel. MSRs are already extremely ambitious undertaking it makes sense to keep it simple, demonstrate the concept, cost and materials can take it then evolve the concept from there.
     
  5. Aug 12, 2015 #4
    I read Transatomic's whitepaper a while ago and found their concept exciting and fascinating. But as the information rattled around in my head, some questions began to emerge. Transatomic doesn't have a forum where members of the public can quiz them on arcane technical minutia; so I figure the PFs would be the next best thing.

    I am satisfied that MSRs are immune to catastrophic meltdowns and other such typical failure modes that affect pressurized, water-cooled designs. Most of by questions and concerns about MSRs relate to fuel-salts leaking. The fuel in a conventional rector is in solid form and encapsulated in zirconium. There is little risk of it leaking (except in severe over-heat conditions). Everything is relatively clean. In contrast, a molten-salt rector has ferociously radioactive goop slopping around everywhere. It would seem to me that the central theme of a concept design would be to minimize all structural and mechanical complexity relating to containing the fuel-salt so as to minimize the chance of very-hard-to-repair issues. Pumps, valves, pipes, pipe fittings, seams, and anything with a large surface-area-to-volume ratio ought to be avoided to the extent possible. This is why I think that the freeze plug concept is brilliant. More than just a passive safety feature, it keeps the design simple.

    So those heat-exchangers, external to the core vat, draw my critical eye. Wouldn't it be safer and simpler to just use the walls of the pot to transfer heat to the secondary coolant? They could avoid a great deal of risk by keeping it as simple as possible. The circulating pumps could be left out and replaced with natural convection. This would also side-step the problem of noble metals plating out of solution and plugging-up the heat exchangers. All the noble metals that a reactor could produce in its life time could coat the inside of the reactor vessel without adverse effects (if anything, it'd help with corrosion resistance). And I remind you that cracks in heat exchangers is only of the most common fatal problems for LWRs and have brought down many a nuclear plant-- and they're not even dealing with radioactive fluids.

    What is the plausibility of doing it like this? I did some back-of-the-envelope math assuming 100m2 of surface area, 3cm thickness, and thermal conductivity of 20W/m-k for Hastelloy-N and the stated power of 1250MW. It came out to a temperature differential of 18,750 degC. Yikes, did I do that right? If so, then I can understand why they went with full-blown proper heat exchangers. But surely there would be ways to integrating this function with the core such as to largely eliminate the possibility of leakage. Even significant design modifications would be worth it in my opinion.

    I'd be interested in hearing what a real nuclear engineer has to say about this.
     
  6. Aug 14, 2015 #5

    etudiant

    User Avatar
    Gold Member

    Does your calculation factor in the conversion losses, seen that a 1250 MW reactor will be producing about 3500 MW thermal?
     
  7. Aug 14, 2015 #6
    They're designing a medium sized reactor intended to produce 520MW net electrical power from 1250MW thermal. The higher than typical conversion efficiency comes from the 650C outlet temperature.

    http://transatomicpower.com/white_papers/TAP_White_Paper.pdf [Broken]
     
    Last edited by a moderator: May 7, 2017
  8. Aug 15, 2015 #7

    Astronuc

    User Avatar
    Staff Emeritus
    Science Advisor

    One would have to calculate the surface area of the walls of the vessel and determine the heat flux, and how much heat can be transferred to the working fluid outside of containment. This scenario would not be efficient for moving heat from the central or interior region of the core, since it has to be transported radially through the outer region of the core then through the vessel. The interior of the core would necessarily be much hotter than the periphery, and the power generation falls off with the neutron flux, which necessarily must fall of toward the containment vessel.

    In addition, one must connect the core to a chemical processing plant in order to remove the fission products, so one might as well include the heat exchanger in the external process system.

    As for the core/reactor internals and pressure vessel, one will have to consider the neutron irradiation of core-internal structures which will degrade the material through radiation damage and transmutation, in addition to whatever chemical and physical reactions (erosion/corrosion) take place between the salts and the structural alloys.
     
  9. Aug 15, 2015 #8
    Well, of course the fuel salt wouldn’t be sitting still; there would obviously be need for some circulation.

    You mention neutron flux. That's something I hadn't thought of. If circulation would be driven by convection, upward through the core and downward along the walls of the reactor vessel, there would necessarily be a variation in density of the fluid. This would affect criticality. It wouldn't be a show stopper; but it would complicate things somewhat. It might even serve as a negative feedback and help safely regulate the power output.

    But even if the ratio of thermal expansion coefficient to viscosity is insufficient for adequate cooling by convection alone, then surely an impeller could provide the rest of the motivation.
    A fair point.
    These factors will have to be addressed regardless of what heat-transfer configuration they use. The only difference would be behaviour of the salt mix under changing temperatures. One of the issues that was discovered at the original MSR experiment was that certain fission products have solubility characteristics that vary by temperature. The result was the these elements came out of solution and plated onto the cooler surfaces (the heat exchangers), creating problems. That's why I mention noble metals coating the inside of the reactor vessel if it was to be used for heat transfer.

    But the key question is whether the sides of the rector vessel, after being made thick enough for safe primary containment, can conduct the heat fast enough.
     
  10. Aug 15, 2015 #9
    To be clear, I don't assume to know enough about engineering to make suggestions as to how they should design their reactor. They've probably forgotten more about engineering than I've ever known. So if I met the two founders of Transatomic on the street I wouldn't say, "Hey, I think you should design your reactor like this." Instead, I would ask, "Out of curiosity, why did you make this engineering decision rather than this other one?"
     
  11. Aug 19, 2015 #10

    etudiant

    User Avatar
    Gold Member

    Heat exchangers are very difficult to execute well, as we again have seen at the San Onofre nuclear plant.
    Yet they are essential, as your calculation shows the reactor will vaporize unless heat is extracted very effectively.
    My hope had been that an inert gas such as helium could be bubbled through the MSR to provide the heat removal, but I believe the problems of managing very hot helium under pressure are at least as great as those associated with molten salt heat exchangers.
    New insights would be very welcome in this space.
     
  12. Aug 19, 2015 #11

    mheslep

    User Avatar
    Gold Member

    I take your point about the importance of heat transfer but this is an molten salt design. If it overheats if will go sub-critical, and if decay heat power is too high it will melt the freeze plug used in all these designs and find a larger surface area to transfer heat.
     
  13. Aug 20, 2015 #12

    etudiant

    User Avatar
    Gold Member

    Thank you for bringing us back to reality. In short, no heat exchanger, no functioning reactor.
     
  14. Aug 20, 2015 #13

    mheslep

    User Avatar
    Gold Member

    That point helps me clarify the problem with the suggestion that the MSR structural vessel be used as the heat exchanger, without need to quantify the parameters of heat exchanger: the function of the two are in conflict. The structural vessel is there to provide highly reliable structural support under all conditions; the exchanger must achieve nearly the opposite, provide as little barrier as possible between primary and secondary.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Transatomic - Molten Salt Reactor (from Question about LFTR?)
  1. Question about LFTR? (Replies: 79)

Loading...