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

Seeking examples reliably safe fission reactor designs

  1. Oct 14, 2011 #1
    I'm writing about emerging technologies the richest people on Earth would be thrilled to have. In fact, its already hit the noosphere. Now I'm writing another appeal to investing in people (this one focusing more on biopersonal impacts than technosocial development) and wish to publish it with a concise explanation of intraplanetary post-scarcity.

    What are some reliably safe designs for fission reactors? I'm particularly interested in thorium reactors as they cannot melt down1.

  2. jcsd
  3. Oct 14, 2011 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    The article in Physorg is about accelerator driven systems.

    As for melting, thorium reactors certainly could meltdown.

    Any fission system produces fission products, and fission products generate decay heat. If the heat is not removed via a cooling system, then the fuel material will melt.

    Coolability is an inherent part of a reactor design. Some designs however can tolerate higher temperatures, e.g., reactor using TRISO fuel (with alternate layers of pyrolytic carbon and SiC or ZrC) in gas cooled environment driving a Brayton cycle. Thermal efficiencies increase with temperature, but then so does materials degradation in an irradiation environment or high temperature water/steam.

    Thoria is desirable for its high thermal conductivity and higher melting point than UO2, although thoria systems would contain U-235, Pu-239 or U-233 which are fissile.

    Accelerator driven systems could in theory produce fast neutrons for fission U-238 or Th-232, but also transmute longer-lived fission products to shorter-lived fission products. The feasibility however is yet to be determined.
  4. Oct 14, 2011 #3


    User Avatar
    Science Advisor
    Gold Member

    The ESBWR design utilizes natural circulation and large accumulators, giving it 100% passive 72 hour safety capability, even under station blackout conditions.

    There are also designs for sealed, passive reactors which use temperature feedback for control, which is also inherently safe.
  5. Oct 14, 2011 #4


    User Avatar
    Gold Member

    There are liquid nuclear fuel designs, but no working implementations aside from a couple experimental reactors in the 1950s and 60s. In other words they operate in an 'already melted' state. Should the primary cooling mechanism of such a reactor stop or fail for some reason so that the fluid temperature continues to rise, the reactor releases the hot fluid by means of melting a fail safe freeze plug and then into a dump tank where the fluid can cool down safely. The dump tanks and freeze plug are shown in blue in this diagram:
  6. Oct 15, 2011 #5
    The sealed reactor designs sound safest as that suggests to me they don't leak radionuclides. Please tell me more about this.
  7. Oct 15, 2011 #6
    Why are those reactor designs so safe? I need to express this as clearly as possible.

    I'm thinking if current radionuclide storage techniques contain nearly all the hazardous materials, overscaling the thickness of the containment material will eliminate risks posed to humans by such.
  8. Oct 16, 2011 #7
    I need the math on how fast dangerous measurements can rise and the relative speed with which they can be brought under control.
  9. Oct 17, 2011 #8


    User Avatar
    Science Advisor

    Best information I have seen for ESBWR is the design certification descriptions at
    http://www.nrc.gov/reactors/new-reactors/design-cert/esbwr.html. NRC site has other design certifications for AP1000 (in progress), ABWR and is in the process of reviewing others.

    Math may be limited as much of the detailed design is considered proprietary. But if you need that level of information you probably need to team up with some technical experts to summarize the technical work. Even then, for the number of different designs, that could fill volumes.
    Last edited by a moderator: Apr 26, 2017
  10. Oct 18, 2011 #9


    User Avatar
    Science Advisor
    Gold Member

    The greatest risk factors for current commercial nuclear power plants is loss of coolant (e.g. by piping break) or loss of heat sink (e.g. by loss of feedwater flow due to station blackout). Since the ESBWR is designed to be cooled by natural circulation, even a total loss of power would not prevent the core from being cooled. Furthermore, the large accumulators allows the core to continue to be cooled by passive/gravity driven systems which do not require operator intervention or power to operate valves/etc for 72 hours, and minimal intervention allowing indefinite passive cooling.

    The sealed reactors are small and therefore do not rely on active cooling to prevent damage, and are further protected by inherent nuclear design which precludes the possibility of a power excursion.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook