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

Why isn't tungsten used in nuclear reactors?

  1. Oct 3, 2018 #1
    Me again, with another potentially ignorant nuclear science question:

    Why isn't tungsten used to prevent meltdown in nuclear reactors?

    If tungsten has a higher melting point of tungsten is almost 6200 degrees Fahrenheit, and nuclear meltdown happens when the uranium fuel is some 5200 degrees, why not line the bottom of reactors and containment vessels with tungsten in order to prevent melt-through and subsequent contamination of groundwater underneath the facility?

    (Unless of course the meltdown can get hotter than 5200 degrees, but I couldn't find the actual highest temperature of a nuclear meltdown; just the melting point of the uranium fuel. Second question, what is the highest temperature nuclear materials used in reactors can reach?)
  2. jcsd
  3. Oct 3, 2018 #2


    Staff: Mentor

    There are many properties of materials that make them suitable or not suitable for use in reactors. Strength, corrosion resistance, soluability ductility/brittleness, and more. High energy radiation also alters material properties, especially embrittlement. Trace amounts of natural contaminants in materials can also disqualify them.

    Rest assured that tungsten and every other potentially interesting material was considered. If it was rejected, then other materials were better.
  4. Oct 3, 2018 #3
    That is still not the maximal temperature of damaged fuel. If it stays contained (without cooling to balance heat production) in narrow space the temperature will keep climbing indefinitely.
    The actual approach is to dilute and scatter the molten fuel (google up 'core catcher').
  5. Oct 4, 2018 #4
    That is very interesting. Thank you! I have not seen that in any of the reading I've done so far.
  6. Oct 4, 2018 #5


    User Avatar

    Staff: Mentor

  7. Oct 5, 2018 #6
    Thank you for your answer! However, I was looking for more of a specific reason why it has obviously been rejected as a material. I cannot possibly claim to know more than trained nuclear physicists, and I am aware that there must be a reason why it is not used, that has been well-researched and tested. I'm more curious about the specific reasoning as to why. I cannot find a clear answer to that question on my vague google searches, so that's why I've come here. :) Thank you!
  8. Oct 5, 2018 #7
    That's fascinating! Now my question is whether or not this has been tested and proven to work. How would one test a system such as this without initiating a meltdown? (Which honestly seems counterproductive, and the risks involved seem to outweigh the reward...) Is this system mandatory in 'new' nuclear reactors?

    Which leads to another question, are there any instances of scientists initiating meltdowns on purpose, in order to study the process? How would such a procedure be performed? How could it be performed safely?
  9. Oct 5, 2018 #8


    User Avatar
    Gold Member

    Apparently, it has been done, and it doesn't look like it was done safely!


    I did not know that.

    My guess is cost.
    Nuclear reactors are not designed to fail.
    They are VERY nasty things, when they do.

    BTW, I can really appreciate your inability to find answers to these questions. Although I'm somewhat versed in nuclear technology, I'm having a very difficult time getting answers, via google, to answer your questions.

    One thing I haven't seen mentioned is something called "decay heat". Though, Rive kind of inferred it:
    Same thing.

    From some back of napkins maths I just did, a small 1000 Mw nuclear reactor core, with a solid volume of about 1/2 m3 would still be generating 30 million watts, an hour after it had been shut down.

    Image I interpolated the volume from:


    1/2 m3 is about the size of a tiny refrigerator.
    30 million watts, is, a lot.
    I'll let you do the maths, if the core had the thermal properties of say, steel, as to what the temperature would be, after an hour.
    Too much maths for me right now.
    Just had lunch, and it's time for my nap.

    ps. I considered the original question to have been mostly answered in post #2. ie. It's VERY complicated.
  10. Oct 6, 2018 #9
    I think you are right with this. As it seems, tungsten can be considered a proper 'nuclear material' since it is in use for long (google up 'demon core'). I could not find any negative effects, like -for example - for Cobalt, which is practically banned from reactors, despite being a well known useful component of many alloys. So, I too think it'll be about the high cost for not enough benefit.

    If there is no cooling, then a meltdown cannot be contained in the RPV, regardless the materials used. If there is cooling, then any decent steel can can do the trick. So let's go with the steel can, and add more concrete to the primary containment... I guess.
  11. Oct 6, 2018 #10
    Good idea, it's possible to invent the system of tungsten tubes under reactor, which will separate the melted core into several fluxes, subcritical each of them, and direct them into different cooled places under reactor.
    World production of tungsten, about 40000 tons/year, is not very big but probably will not make a big bareer for such project.

    Another potentially possible way of using tungsten is using W184 isotope for shells of fuel rods in the core. It has capture cross section about 0.5 barns compared to ~20 barns of natural tungsten isotopes mixture.
  12. Oct 6, 2018 #11


    User Avatar
    Gold Member

    Wow! I read the wiki entry on it. Hilarious and sad, at the same time.
    What about silver? It has the best thermal conductivity I could find. You just need a whole lot more money, and assume that there is going to be an accident.
    After my nap yesterday, watching 6 hours of "Rick and Morty" reruns, and a full nights sleep, I researched how many reactor vessels were breached. I only came up with Chernobyl and 3 of the Fukushima Daiichi vessels.
    I was surprised to see that the Three Mile Island vessel was never breached, even though 19,000 kg of the core melted! [ref]
    Good lord!

    One strange thing I found out about the Three Mile Island "Corium" was that it was 70% Uranium by weight. [same ref as above]
    I guess I've been out of the industry too long, as that sounded like quite a healthy concentration to not be critical.

    I was going to research that some more at our Japan Earthquake: nuclear plants thread to find what I was not understanding, but it's over 7 years old, and has nearly 16,000 comments. :oldsurprised:

    Complicated, is an understatement, IMHO.
  13. Oct 9, 2018 #12
    If you want to consider why or why not something was used, the first consideration often is money.
  14. Oct 9, 2018 #13
    I think this is it. The vessel is designed to maintain the coolant pressure, and keep the fuel submerged in the coolant. While it is much nicer post-accident cleanup if the fuel remains in the vessel (as TMI) it wasn't a design criterion for the vessels.

    Regarding ex-vessel "core catcher" this was considered by the US regulators in the 1960s. David Okrent's book goes into this in some detail. See NRC ADAMS website, search for accession number ML090630275, "On The History of the Evolution of Light Water Reactor Safety in the United States".


    By the way, this document is fascinating if you have an interest in the history of nuclear regulations in the US. But be warned it is over 1000 pages.
  15. Oct 12, 2018 #14

    Mark Harder

    User Avatar
    Gold Member

    I was thinking about using W for small rocket nozzles. In researching its properties I found that W is very brittle and therefore difficult to work unless it is very pure. In order to spin, bend & forge tungsten it needs to be something like 99.99% pure. In contrast to most industrial metals, that's a pretty unforgiving specification. Making a vessel out of many, many tons of a 99.99% (again, that's just an order of magnitude guess) elemental metal has got to be a very expensive proposition. Also, W has a density of 19.35 g/cm^3, slightly greater than gold, so unless a thinner W PV is sufficiently strong compared with steel, etc., you would also need stronger materials for all the structures that support the PV. There are alloys of W that melt at higher temperatures and resist the tendency to creep under tensile force at high temperatures. They are used to make impeller blades in military jet engines. The alloy is a mixture of tungsten and rhenium. The latter is one of the rarest natural materials in the periodic table, so unless you have a DOD contract, you're not going to be using rhenium to build big pressure vessels.
  16. Oct 13, 2018 #15


    User Avatar
    Staff Emeritus
    Science Advisor

    Tungsten isn't used for design reasons and cost. While tungsten has a high melting point, it is subject to corrosion and embrittlement.

    PWR and BWR pressure vessels have numerous penetrations in the bottom part of the pressure vessel. PWR have instrumentation systems, e.g., thermocouple tubes and flux thimbles that allow thermocouples and neutron detectors to travel into the core. Some systems are designed to insert instruments from the top, which reduces or eliminates need for penetrations. BWRs on the other hand have control rod and detectors inserted from the bottom of the core. The penetration tubes and lower core structures are composed of stainless steel, usually a type of 304 or 316, but some could be 347 or 348. The pressure vessels are type SA 508 Cl 2 or SA 533 B, and the inner surface is clad with an austenitic stainless steel. Adding a layer of W-alloy (e.g., alloyed with Re or some other elements) to the pressure vessel inner surface would be rather impractical. One would have to protect the tungsten alloy from corrosion by the coolant and interaction with the other structural materials, and deal with differences in thermal expansion as the reactor vessel expands to operating temperature and contracts to cold shutdown conditions.

    The core support structures are cast stainless steel. The upper and lower nozzles of PWR fuel assemblies and upper and lower tie plates of BWR fuel assemblies are made of wrought or cast stainless steel, usually a 300 series, e.g., 304/316/347 or derivatives. Casting equivalent is often CF3/CF3M.

    The actual maximum temperature of corium is complicated and can only be estimated on a case-by-case basis. The Wikipedia article makes a good estimate, but it's up to the melting point of UO2. The maximum temperature will depend on the decay heat, composition of the melt, porosity and how much coolant is available. Furthermore, since the melting point of stainless steel is ~ 1375-1400 C and Zr-alloys ~ 1850 C, that would pretty much limit the temperature of the melt, although it could go higher to melting point of metal oxides if steel and Zr-alloys react (oxidize/corrode) in high temperature water.

    The melt temperature does not increase indefinitely, but is limited by what supports and/or interacts with the melt. One can find many simulations and experiments by searching on "Corium simulations" or "corium experiments", or "reactor severe accident analysis" or "experiments".

    The original design of LWRs calls for emergency core cooling systems. Modern reactor system designs have more passive features.

    In the case of Fukushima, we're still trying to learn what happened there. Clearly there was an chemical oxidation reaction with whatever water was present, from which produced the hydrogen that exploded. Clearly the cores had insufficient water in the system, or the core was so hot in stagnant steam/hydrogen that the metals simply reacted and disintegrated (some believe melting).

    Getting back to a layer of tungsten, with a mass of molten core sitting on top, the heat would transfer to the softer steel under the W-alloy layer, or melt the stainless steel penetrations, and possibly the W-alloy would react with the steels. So some layer would have to be present, e.g., ceramic. So one is still left with differential thermal expansion and other issues.

    BTW, tungsten is used in some control elements. See US patent 8537962 B1.

    W-184 would make an interesting reflector.
  17. Oct 25, 2018 #16
    Thank you so much! This is very informative. Also thank you for the terms to search. I think half of the research problems that I am having are because I'm not sure how to adequately word my searches to get the best answers. Thank you!
  18. Nov 7, 2018 #17
    I think the most critical problem with tungsten in reactor would be corrosion resistance. Tungsten form oxide which is powdery and do not offer corrosion resistance above 200C temperature.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?