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

More efficient nuclear energy?

  1. Jul 28, 2012 #1
    Isn't there any better/more efficient methods of generating electricity from fission/fusion than basically using the heat to boil water to spin a turbine? I know that that there are different generations of nuclear reactor designs (with each more efficient than the last) that get better and better at extracting the heat from the reaction. But at the end of the day, for pretty much every design I've seen, you still use that to boil a liquid, which spins a turbine. It seems too much like just a coal plant in that regard. Or am I being too critical, and that the current paradigm is already very efficient at turning the nuclear energy into electrical energy?
  2. jcsd
  3. Jul 29, 2012 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    Currently, there is no better and reliable way to produce electricity than boiling a liquid and driving a rotating turbine - with nuclear energy or otherwise. The higher the temperature, the more efficient the energy conversion, however, higher temperatures usually come at the expense of material performance with erosion/corrosion and/or creep being the limiting factors. In the case of nuclear systems, containment of fission products becomes an issue as temperatures increase.

    The steam Rankine cycle has been the traditional method of producing electricity with large scale systems (100s of MWe to 1+ GWe) for decades. Older nuclear plants had efficiencies of about 32-33%, but modern ones have efficienies approaching 37% - mostly due to more efficient turbine (blading and seals) design. Nuclear plants are considered wet steam plants because they do not use superheated steam - because of safety concerns, and well as plant performance issues. Fossil fired plants can use superheat, and some plants achieve efficiencies in the range of 34-38%. Plant employing supercritical or ultrasupercritical thermodynamics cycles may achieve conversion efficiencies of up to 44%.


    Some plants may use combined cycles, e.g., Brayton gas-fired cycle (based on an aero-derivative gas turbine) and steam Rankine cycle (heated by the exhaust of the gas turbine), and efficienies approach 60%. Nuclear Brayton systems have been problematic, and combined cycles plants would also be problemetic from the standpoint of heat exchanger preformance - the heat exchanger would have to maintain close to 100% reliability in keeping the Brayton system and steam system separated.

    In fusion, the current plan is to use the thermal energy to drive conventional steam cycles. Ideally, MHD or direct conversion could be used, but appropriate configurations in conjunction with plant design has proved elusive. In fact, viable fusion conversion systems have proved elusive as well. In direct conversion, the ions and electrons are separated in a magnetic field and they provide a direct current to the load. Theoretically, efficiency could approach 80+%. However, there are practical engineering challenges.

    Thermionic concepts have been considered, but they too are complicated and much less reliable - as well as expensive.
  4. Jul 29, 2012 #3
    Generally speaking, LMFBR and HTGR's efficient are higher than PWR and HWR.
  5. Jul 30, 2012 #4


    User Avatar
    Staff Emeritus
    Science Advisor

    Please provide examples.

    Certainly, there were advanced HTGRs planned for the US about 40 years ago. They were rated at 42% efficiency. None were constructed. Fort St. Vrain did operate, but they had significant operational problems, and the plant was shutdown in 1992 after 15 years of operation.

    http://en.wikipedia.org/wiki/Fort_St._Vrain_Generating_Station (there is incorrect information in the article)

    The other US HTGR, Peach Bottom 1, operated from 1966 to 1974.

    Similarly, the US does not have an LMFBR in commercial operation.
  6. Jul 30, 2012 #5
    How does the direct conversion differ from that of an MHD generator, and what are the problems in implementing them? And is it possible to implement them in fusion reactors? Because I find the notion of using those more appealing than the current paradigm.
  7. Jul 30, 2012 #6


    User Avatar
    Gold Member

    There are creative concepts for direct conversion, usually involving electrostatic fields to capture the energy of the charged particles emitted from the nuclear reaction. Of course this requires careful selection of the nuclear process used to make sure it produces as few neutral particles as possible. The engineering of the fields to serve as power sources has of course never been demonstrated, even on a small kilowatt scale afaik.

    MHD is an entirely different approach to produce electricity, the heat of a gas ionizes a carrier material, usually cesium. The negative and positive ions are then collected by electrodes within a powerful magnetic field, creating nearly direct heat to electricity conversion. The material requirements however are severe and the approach has never been used commercially afaik.

    Both of these concepts would require major development to bring into being and it is likely that both would need so much support technology that the wonderful simplicity of the core idea gets lost in a welter of expensive and unreliable ancillary essentials.
  8. Jul 31, 2012 #7


    User Avatar
    Gold Member

    Or otherwise, as in including non-nuclear? I think that statement would be hard to defend without picking some parameters and discarding others under the cloak of "better". There must be some ~100GWe of gas turbine Brayton installed in the US. "Better" for Brayton turbines would include better efficiency, much higher when including use of combined cycle on the Brayton rejected heat; less use or reliance on a major source of water*, and better power density. I'd concede cost advantages and perhaps reliability to Rankine boilers-steam turbines, though I'm not sure about reliability as all those gas turbines clocking miles five miles up in the air seem to be fairly reliable.

    Exactly. Sounds like better to me.

    *Particularly relevant is the early closing of NJ's Oyster nuclear reactor on the Atlantic coast. It's directly water cooled from the local bay, but NJ officials complained Oyster was doing too much harm to sea life drawn from the massive water intakes, demanding Oyster switch to evaporative cooling towers. Oyster's operator has said no thanks and will close prematurely.
    Last edited: Jul 31, 2012
  9. Jul 31, 2012 #8
    Thermodynamic efficiency is important, and does of course have an aesthetic appeal, but if you are in the electricity generation business, it is the optimum combination of efficiency, fuel cost, and capital cost that's of interest. If you are a consumer of the electricity, it may be the optimum combination of those factors along with others (e.g., minimum emissions, fish kill, etc.) that you find more aesthetic.
  10. Jul 31, 2012 #9
    Yes, none of them were constructed. But some were designed.

    CRBRP's design efficiency was 35.9%, what is from page 24 of THEMAL ANALYSIS OF LMFBR.
  11. Jul 31, 2012 #10
    So it sounds like the improvement in the reactors is more of an incremental thing based on the current way of doing things. Since that even if a viable, efficient method for direct conversion or MHD was developed tomorrow, it would take years or maybe even decades to get a commercial reactor out of it given all the testing that would need to be done.
  12. Jul 31, 2012 #11


    User Avatar
    Staff Emeritus
    Science Advisor

    The focus on current commercial nuclear plants has been increasing capacity factor and availability (with longer cycles and fewer refueling outages, and reduction of unplanned outages), materials reliability and corrosion mitigation, plant uprate, and life extension. Some plants have installed new turbines and realized 2 or 3% gain in conversion efficiency.

    A number of PWRs have had to replace the original steam generators because the Inconel 600 tubing failed prematurely, lasting less than 30 years rather than the planned 40 years. Other plants have implemented Zn-injection and/or reduced primary circuit temperature in order to extend the life of current steam generators.

    Some early examples - ftp://ftp.eia.doe.gov/features/steamgen.pdf

    Advanced energy conversion techniques such as direct conversion or MHD require something different than current LWR technology.

    Some Gen-IV systems are designed for higher temperatures, SCWR or LMR or MSR. But those are only designs, and all have significant challenges with respect to materials performance (e.g., corrosion, erosion and creep) at high temperature.
  13. Aug 1, 2012 #12
    Why not design steam generators with more tubes, but slower flow?

    What's the point in trying to push them to limits of what materials can withstand and then be haunted by problems of material degradation?

    It's not a spacecraft where every pound matters, right?
  14. Aug 1, 2012 #13


    User Avatar
    Staff Emeritus
    Science Advisor

    AT the time it was selected, Inconel-600 was considered accepted. It had been used in other applications. However, it had not be used necessarily in the environment such as that encountered in PWRs - boric acid and LiOH - and pH < 7.0 (neutral) at 300-330°C. In Europe, Incoloy 800 has been the prefer SG tube material, while in the US and Asia, Inconel 690 has been used to replace Inconel 600.

    Lower flow rate would not work in the core since heat transfer would suffer. Enthalpy rise and coolant temperature are critical factors in nuclear fuel performance in LWRs. Total flow rate and enthalpy of the steam determine the power available to the HP and LP turbine set.

    In the case of PWRs, primary water chemistry has been optimized over that last 3 decades. Now pH is above 7.0 to the extent possible, and 7.4 is a typical target for much of the cycle. Some plants add Zn to the coolant to mitigate corrosion of SG tubing and control crud deposition (and activation) in the core.
  15. Aug 1, 2012 #14


    User Avatar
    Gold Member

    Yes for generators include efficiency, O&M, capital as you say. There are others factors. *Generators* also have an interest in reliability, and not just of the new plant but of its impact on fleet reliability; time to build (construction & approval), and environmental footprint. Yes these are all connected to O&M and capital costs, but the large generator has to look at long term and fleet wide service. As a generator yes I can build a 93% cap. factor large nuclear plant, but what's my plan when it inevitably goes down for some months or a year (flooding, minor earth quake) for extended inspection? How does that impact compare with building multiple and distributed smaller combustion plants? Yes I might get a new coal plant permitted, but what legal harassment or other government interference might I expect half way through the plant's life, causing me to suffer stranded costs?
  16. Aug 1, 2012 #15


    User Avatar
    Science Advisor
    Gold Member

    You could but then the reactor would have to have a lower power density and thus less economically efficient. Pounds (lbs) do not matter but Pounds (money) does.
  17. Aug 1, 2012 #16
    Yes yes and yes... My point is that thermo-efficiency (the OP subject of this thread), while interesting, is not the only consideration in selecting the 'best' design. There are other attributes to consider. Many others, as you rightly point out.
  18. Aug 1, 2012 #17

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    amen to last 3 posts

    in real world , efficiency = energy out / energy in ;

    in surreal world , efficiency = revenue / expenses ;
    Last edited by a moderator: Sep 25, 2014
  19. Aug 2, 2012 #18
    I meant "lower flow rate through individual SG tubes": if you have more tubes in SG, then you don't need to pump water as fast through each tube to achieve the same flow rate through SG.

    Will this help in reducing vibration and wear?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook