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

Why can’t decay heat be harnessed to safely shutdown a nuclear reactor?

  1. Aug 13, 2011 #1
    Why can’t decay heat be harnessed and used as an energy source to safely power down/cool a nuclear reactor?

    I have been wondering about this since the reactor incidents in Japan as it appears a tremendous amount off energy must still be dissapated after the shutdown of a nuclear reactor in order to prevent damage to the reactor and a potential disaster. It appeared that the Fukushima Daiichi nuclear reactors survived the 5th largest recorded earthquake on earth remarkably well and initiated normal shutdown procedures. It was the fact that the tsunami later damaged the backup power system which relied on diesel generators for cooling, resulting in a cascade of failures and a core meltdown in three of the reactors. I feel that nuclear energy is a clean source of power and that it can help solve our dependence on imported fossil fuels, promote national security, as well as provide virtually no CO2 emissions. On the other hand, plants should be designed to withstand extreme events, even if they are of a very low probability. In the Japan case, ancient stone markers warned of tsunami risk at levels well above the Fukushima backup generators and the plant designers/operators disregarded something left between 500-1000 years ago. I know that many facilities such as this are built, they are meant for an operational life of between 40-75 years. The consequences of something going wrong at a facility such as this are simply too large to design for on an operational life scale. Had a few minor circumstances in this situation unfolded differently, we could have been looking at the evacuation and resettlement of Tokyo for between 100-300 years due to radioactive contamination.

    Most large engineering/industrial disasters include "the human factor" as a key component of how they unfold. These includes the Chernobyl disaster, Deepwater Horizon oil spill, Exxon Valdez, Challenger disaster, sinking of the Titanic, and many others. Had a few simple decisions been made differently, none of these would have happened. Unfortunately it is very rare when the human factor is recognized as a main factor in preventing a major catastrophe. Perhaps the best known recent incident is the U.S. Airways Flight 1549 or "Miracle on the Hudson" incident. An experienced airline pilot made some fateful split-second decisions after realizing that all his engines had been disabled by a collision with a flock of geese. It was enough of a miracle that he safely "ditched" on the Hudson and no passengers on the aircraft died. This also occurred in one of the most populated areas on the planets and could have easily resulted in many hundreds or thousands of deaths on the ground had this plane not been ditched successfully on the Hudson but instead came crashing down in a populated part of New York or New Jersey. The human factor likely also played a major role in preventing the Fukushima incident from being far worse.

    As an engineer and a scientist, I dislike getting information on important topics through normal news outlets that like to sensationalize and oversimplify stories. I understand that I am not a nuclear engineer so maybe this is a dumb question but I have dealt with lots of disasters including Katrina and know that failures of the power grid over an extended period could result in the loss of backup cooling due to diesel fuel running low and such. I understand that backup cooling systems at U.S nuclear reactors are required to be able to operate for 72 hours. It seems something more robust and redundant should be used as an earthquake or solar flare could disable enough of the power grid to not allow for the safe shutdown of many nuclear recators.

    It is my understanding that the typical reactor will produce between 5-7% of its rated output in decay heat due to the radioactive decay of fission byproducts after shutting down and decay down to around 1% within a day or so. I understand that the amount of heat generated depends on the length of time the fuel has been in use and undergoing fission so older fuel will have a larger and longer lasting decay heat concern. I understand the heat generation drops quite rapidly as the short lived isotopes decay but that longer lived isotopes continue to decay and generate a significant amount heat so that cooling is required for a very long time (5-10 years) after the spent fuel is removed from service.

    I looked up the operational rating of several nuclear power plants in the U.S. and most tend to range between 1000-1200 Megawatts (MW) of power, which is quite a large number. When one of these shuts down, decay heat should be generated in an amount around 50 MW (or more) immediately after shutdown based on the 5-7% heat of operational output. 50 MW is an immense amount of power (enough to power about 50,000 average U.S. homes) and I would think this would well exceed the rated output of even the largest (or a bank of) diesel generators used to provide backup power to cool a shutdown nuclear reactor.

    My question is why this tremendous amount of energy cannot be harnessed and used to generate power that could be used to safety shut down and cool a nuclear reactor? It seems there is plenty of heat to lead to a complete core meltdown and/or fire long after the primary fission reaction is shut down. Why can’t this heat be used to generate power, whether it be electrical or mechanical, in order to run pumps and such to cool the reactor during shutdown? Why couldn’t one of the steam turbines be run to generate power to run the pumps? If the main turbines are too large to run on such a reduced output, could a smaller turbine be used for backup purposes? How about running the pumps directly and mechanically without any electric generation via a turbine meant just for this purpose? I like to keep things simple as there is less to go wrong so a purely mechanical pump might be in order. How about a thermocouple system? I know that radioactive decay heat is used to power space probes such as Voyager 1 and 2, among others, in this manner and such but don’t know how it would work on such a large application. Even if decay heat cannot produce enough power, can it not provide some power and reduce dependence of batteries or diesel? If nothing else, it could reduce the rate at which batteries and/or diesel are depleted and buy time to solve the underlying problem.

    As decay heat drops over time, potential power generated from it also drops, but so would the cooling requirements. Pumps would not be able to be run at their maximum rating but is this a bad thing after most of the short lived isotopes have decayed? I am not an expert so maybe decay heat can remain dangerous even if it isn’t enought to generate a meaningful amount of power. Is it like my electric stove. Sometimes I turn it off right before the food is done and let it cook with the residual heat. Eventually it cools off to where it can no longer cook but would still be dangerous to touch. I know this is very simple but is it a good comparison?

    If decay heat cannot effectively be used to shut down a nuclear reactor, why can’t the reactor be reduced down to a down to a "fail safe" or “idle” mode where it generates just enough power to run the emergency cooling systems? It could be run this way indefinitely and let some of the short-lived isotopes generated during full power operation decay over a period time before reducing power further or shutting down completely once enough short-lived isotopes have decayed. Why is this not done in emergency shutdowns where the reactor infrastructure is left intact but external power is lost?

    All it takes is one unforeseen disaster to knock out external power at a nuclear plant and it seems this might be a solution or at least part of the solution to the decay heat issue. I have been reading about solar flares and their ability to fry large electrical transformers that are key to large parts of the power grid. I understand that we are entering a very active solar cycle and there is some concern one of these flares could knock out a large part of the grid for an extended period and it might takes months or longer to restore some parts of the power grid due to the destoyed transformers. What would happen to a nuclear plant in such a situation where external power is lost for a very extended period of time?

    I have never heard this question asked and did extensive searching online on this subject, only to find nothing. Engineers and scientists realized that the incredible amount of power contained within the atom could be used for peaceful and beneficial purposes even before the first atomic bomb was dropped. It seems that the decay heat generated after a nuclear reactor is shutdown could be harnessed and used to cool a reactor instead of being viewed only as a destructive form of energy that must be released in order to prevent a disaster.

    Thank you.
  2. jcsd
  3. Aug 14, 2011 #2
  4. Aug 14, 2011 #3
    Old BWRs, that is. I have been told by a veteran BWR designer that one reason why they gave up diversity and concentrated on electrically driven safety systems and ensuring the emergency power in the plants designed in late 70's and 80's was the somewhat poor technical reliability of the turbine driven pumps. So in spite of being in principle very tempting way to increase system reliability, they did not turn out very well in the PRA sense.

    New BWR plants typically have passive heat removal systems (=isolation condensers) to remove dependence on electricity supply in plant blackout situations, which may be a problem in the BWR:s designed in late 1970's to late 1990's.

    I am not an expert in mechanical engineering, but I have a hunch that a duplex pump in a pressure vessel blowdown line might from the reliability point of view be a good candidate for backfitting old NPP:s to endure total plant blackout situations. However, there would still remain the problem of transferring heat away from the containment, and thus need for (filtered!) venting.
    Last edited: Aug 14, 2011
  5. Aug 14, 2011 #4


    User Avatar
    Gold Member

    Was not the problem the absence of fresh feed water for the system?
    If the condenser gets too hot and there is no makeup water, because all the feed pumps are out, no steam engine, whether turbine or steam piston driven duplex pump will work.
    Afaik, the AP 1000 design has a built in gravity feed reservoir to address this. It would provide for several days of cooling before needing replenishment.
  6. Aug 14, 2011 #5
    Getting sufficient amount of feedwater to the secondary side of the isolation condenser is one thing, if such a system exists in the plant. Steam driven pumps could in principle be used in this task also, but also e.g. firefighting systems etc., since there's usually some amount of time to get things arranged due to the relatively large water volume on the secondary side of the IC.

    Another thing is the plants that do not have an isolation condenser at all (i.e. majority of BWRs), where the core will typically uncover within an hour if all power is suddenly lost at scram time. In those cases, there's plenty of water available in the containment suppression pool, if there only was a way to pump it into the reactor. It is this latter type of plants I was referring to with the duplex pumps as one possible backfitting option.
  7. Aug 14, 2011 #6


    User Avatar
    Gold Member

    This is a naiive question - but why can't decay heat be used to create steam to spin the turbines to generate the power to drive the valves and pumps - and therefore provide primary cooling? It seems that the first thing that trips is the turbine.

    At Fukushima the Tsunami would have prevented this mitigating the accident I know, but to my simple mind this seems the obvious way to mitigate a station blackout for a period of time that would exceed the battery backup. Perhaps weeks...

    Edit - please ignore - I have duplicated the OP's question... I should have read it more carefully.
  8. Aug 14, 2011 #7
    The main turbine is rather a delicate beast, and can not be run with the small steam flow that is available following a scram. You would need a separate emergency turbine with all the auxiliary equipment and systems needed for making it run. The turbine driven pumps used in the old BWR:s is one approach to this, driving the pumps directly without need for generating electricity.
  9. Aug 14, 2011 #8


    User Avatar
    Science Advisor


    That is EXACTLY what the operators of Chernobyl were attempting to do when they caused
    the world's greatest nuclear power accident.

  10. Aug 14, 2011 #9

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Catch 22 flag thrown.

    Actually, part of what you propose is not uncommon.
    Most plants have steam driven pumps that supply water to the boilers and, in absence of offfsite power that is how the plant gets cooled down from operating temperature to the point there's not much steam pressure - maybe 300 degF.
    They're big pumps so powering them from steam makes sense, it leaves room on the diesels for other important stuff like battery chargers and seawater pumps and emergency lights and instruments, and some oil pumps to keep the hydrogen in that expensive main turbo-generator ......

    There's another practical point to consider:
    Were your other pumps and valves and lights and stuff powered by that steam turbine , you couldn't cool down below the point of making usable steam. So they are powered by diesels otherwise you couldn't get to cold shutdown.

    keep it simple -
    old jim
    Last edited: Aug 14, 2011
  11. Aug 15, 2011 #10
    A hybrid system would also be possible, i.e. have electrical pumps powered by a steam turbine powered generator, with internal or external diesels providing backup power if the steam turbine becomes unavailable.

    Also, if the RCIC had not only powered a water pump to top up water in the core but also had been able to generate a bit of electricity on the side, perhaps it could have recharged the batteries that its valves depended on? That might at least have bought some time. I understand the RCIC turbine starts and stops based on the water level in the core, so it wouldn't have been able to provide continuous power.

    I am curious when we will find out what happened to the IC of unit 1. Did it run out of water? Was there any plan to top up is water supply using fire engines? It seems odd that it was turned off for several hours, as has been reported. Do we know if that was just one of its two units or both?

    Tepco was talking about setting up a heat exchanger with air cooling either for the reactors (before they found out that they had melted through and were leaking like a sieve) or the SFPs.

    The old VW beetle engine owed some of its robustness to the fact that cooling water can leak or freeze up without anti-freeze, but you never run out of air.

    Maybe having some kind of radiator with a steam-powered fan as a backup to the sea-water based electrically powered RHR would have helped survive the tsunami, considering also that the only diesel that survived in that situation was air-cooled.
  12. Aug 15, 2011 #11
    It would, but I have understood that the main reliability question lies in the steam turbine, which is rather sensitive to the operating parameters of the steam, lubrication etc. I have the impression that an old-fashioned duplex pump could be made more fool-proof in the sense that it would not need any sophisticated control system: just feed it with some kind of steam and it will force a proportional amount of water upstream. If there's too much water in the core, that would not be a serious problem in case of ultimate emergency: just make sure there is water above the core.

    Still, you would need an alternative power source for the low-pressure core injection needed in case of LOCA, as there would not be steam pressure available to power the pumps.

    My thoughts exactly. My recollection from the documentation published by TEPCO is that there would have been enough water on the secondary side to boil off the decay heat for 90 minutes, which does sound rather a short time window. Still, a lot can be accomplished in that time, if there are practised EOPs in place. It would be really interesting to hear the details why they failed to restart the IC after the tsunami, as the IC is something many of the the new BWR designs rely heavily upon in cases of emergency.

    In the late 1970's to early 1980's there was a district-heating passive nuclear reactor concept SECURE (link) developed as a joint Swedish/Finnish co-operation and further developed and marketed by Asea until the Chernobyl accident finally destroyed the market. One of the safety features in that concept was that the reactor was located underground, making it possible to arrange a natural circulation cooling of residual heat via a cooling tower. The same approach had already been used in the Ågesta heavy water suburban reactor that operated in the Stockholm suburb of Farsta from 1964 - 1974.

    If your plant site does not allow to use natural circulation cooling to the atmosphere, then you will inevitably need some active means to circulate the cooling water to the heat exchangers. All in all, I suspect there's a lot that could be learned from these early attempts to improve reactor safety by passive means that could be applied to new plant designs and back-fitting of old ones, where need for improvements is recognized.
    Last edited: Aug 15, 2011
  13. Aug 24, 2011 #12
  14. Oct 19, 2011 #13
    This might be a dumb comment, but wouldn't feeding steam generators cool down your primary system too quickly? A rapid drop in pressurizer level is never something you wanna see, and charging would only cool down your primary even quicker. I think this would mean that whatever water you have in your steam generators is pretty much all you've got. Once you get to a pressure where you can't pull steam off anymore it's pretty much game over.
  15. Oct 20, 2011 #14
    We need 100% passive cooling systems. Anything else is just not good enough.
  16. Oct 20, 2011 #15


    User Avatar
    Staff Emeritus
    Science Advisor

    In a PWR, which uses steam generators, the rate of heat removal would depend on the temperature of the secondary side and the relative flowrates between primary and secondary side. The primary side would be on natural circulation if power to the coolant pumps was lost.

    In BWRs, like those at Fukushima, the steam is produced in the core. Then the problem is getting cooler water to the core.

    That is the approach used in some of the modern Gen-III+ plants.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Similar Threads for can’t decay heat
How Much Heat is Produced by Radioactive Waste?