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

I Spallation source question

  1. Jul 2, 2017 #1
    I read that spallation is the process where you have a high energy proton source (a linac or other particle accelerators for example) and you direct those protons into a metal target which is of specific alloy or whatnot and you get that the protons impacting the material release secondary particles due to nuclear reactions happening inside the bombarded target so you get secondary neutrons (probably other ejected particles also) which you can then use either for science experiments or for controlled nuclear fission, aka you have a neutron source.

    Now what I am wondering and couldn't find after light google searching is whether it is possible to have this process "spallation" only when a target is bombarded with high energy protons or are there some other ways to achieve neutron emission of any sort in a controlled manner? excluding ofcourse already operational nuclear reactors but a neutron source which doesn't come itself from an already started and self sustaining chain reactions like that of a commercial conventional reactor?

    I assume spallation is in part similar to the photoelectric effect where you have photons aka EM radiation above a certain energy (wavelength) threshold which upon impacting a metal plate is capable of releasing electrons?

    why is it so that photons no matter how energetic cannot ejected particles (neurons) fro the nucleus only electrons but a energetic proton beam can do that in the same material target or am i wrong?

    thank you
  2. jcsd
  3. Jul 2, 2017 #2


    User Avatar
    2017 Award

    Staff: Mentor

    Photons can eject neutrons, but how do you generate photons with an energy of several MeV efficiently?

    Accelerating protons is simply the easiest way.
    Nuclear fission reactors work as well, accelerator-based fusion is a third option.
  4. Jul 6, 2017 #3
    Ok , since I got an answer to a similar question from you in another thread it got me thinking.
    Talking about the idea known for many years which is spallation based fission reactor for example, now the ways of spallation that have been suggested have varied from particle accelerators to maybe lasers or other but in terms of the reactor itself, I assume that one would only need a rather small amount of the external spallation neutrons required for reaching criticality in the core because the core itself would be very close to the critical mass requirement just not slightly there correct?
    And by this one achieves a reactor which can be switched off or on faster and safer more like a switch instead of having the full critical mass in the core that is then controlled via neutron absorption and moderator means?

    Although I assume that even such a core does produce decay heat after switch off for some time but bit less than an ordinary reactor would correct?

    In terms of the spallation itself, I was wondering photon upconversion you mentioned in my other thread, Since I read that there are scientific proposals and work done to implement this is the photovoltaic cells, I have a question could a sophisticated and well engineered LED light source with high intensity work with a special material target in terms of creating high energy photons for spallation? I was thinking about this because normally you said yourself that we don't know how to make high energy photons efficiently so we instead accelerate protons for spallation, but if the upconversion thing works somewhat to the degree I think it does, then could a LED or maybe some other light source do the trick, as far as I know some LED colors are more efficient in terms of electrical energy in and lumens out tan others but in general for a specific single color they are all rather efficient.

    Also what is the constraint of how high of an energy photon one gets out after upconversion, is it dependent on the material used for upconversion or the wavelength of the primary light source or it's intensity or rather I assume some combination of all of the above?

  5. Jul 6, 2017 #4

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    It's hard to decipher that wall of text, but if you are asking if LEDs can make gamma rays the answer is no.
  6. Jul 6, 2017 #5
    I do admit my writing style or rather my thought is not always straight forward although while reading through my last once more I believe the general questions are understandable and I would love and appreciate to hear more from you or anyone else.

    Well ok you say LED's can't make gamma rays, sure enough that sounds logical enough, but then I have a question , regarding the photon upconversion process, what then is the wavelength difference once can achieve,
    let's take for example a imagined setup where I have the already mentioned high intensity white LED source shining on a specific material (a doped crystal of some sort etc) now what I read in papers is that the outgoing photons would be of shorter wavelength aka higher frequency, but I can't seem to find what parameters determine the frequency difference between input and output.?

    PS, is it then true that a spallation driven reactor does have a critical mass which is "a bit short" from criticality and hence can be driven by spallation neutrons to achieve criticality and also shut off by the lack of these spallation neutrons? Atleast that's what I gather from the sources I've read through.
  7. Jul 6, 2017 #6


    User Avatar
    2017 Award

    Staff: Mentor

    Right. MYRRHA wants to run at 0.955, or 22 fission reactions per input neutron. Stop the accelerator and the fission reactions stop within a millisecond. A few delayed neutrons will lead to a tiny amount of fission reactions that happen later, but that doesn't contribute notably to the output any more.
    It would be similar.
    No. Upconversion can give you visible light from infrared, or sometimes soft UV from visible light, but it doesn't work in the keV or MeV range.
  8. Jul 6, 2017 #7
    If you're looking for a neutron source, there are a few options. Spallation sources can be tempting due to the promise of many neutrons per proton, but the proton energies of roughly one GeV require enourmous linacs or proton synchrotrons. The spallation neutron source in Oak Ridge, Tn uses a roughly 1 GeV beam of protons which strikes a liquid mercury target. Fission is another source, especially the fission of U 235. It is different from spallation in that fission is started by a neutron of any energy, and is actually more efficient with lower energy neutrons. However, due to obvious budget concerns, these methods are reserved for organizations with many millions of dollars. Fission can be used as an amplification method. Triuranium octoxide can be obtained very cheaply, and sending thermal (under 2.2 km/s) neutrons through it will cause the .7204% natural content of fissile U 235 to undergo fission, releasing on averge three neutrons per fission, which have too much energy to initiate more fission in a thin sample, and will much more likely escape. The obvious concern is the handling of nuclear waste, and that if U 238 captures a neutron, it will form U 239 with decays in a matter of weeks to Pu 239, aka plutonium. For more detail on thermal neutron interactions, this table provides data on cross sections. https://www.ncnr.nist.gov/resources/n-lengths/
    Fusion is another option for neutron creation. Deuterium deuterium fusion will create a 2.45 MeV neutron in half of all successful fusions. Tritium deuterium fusion will create a 14.1 MeV neutron in all successful fusions. The energies required to achieve fusion are surprisingly low. The probability of fusion in a collison can be calculated by the Gamow factor. Using this, at just 46.5 KeV, a pair of deuterons has a 1% chance of fusion upon collision. This might even be a viable source for you, considering that the "cyclotron kids" build a cyclotron capable of creating 2 MeV protons with less than ten grand of funding from jefferson laboratories. I wouldn't be surprised if someone made a fusion capable deuteron cyclotron on a budget of ten, or even as low as three or four grand. I plan on doing this myself.
    The photon method is basically not worth considering, as the energy requirement is not that of intensity, but of energy per photon, or frequency. To cause neutron emission in beryllium, a 1.665 MeV photon is the minimum. And beryllium is the easiest to split! The runner up is deuterium at 2.225 MeV. Creation of such gamma rays occurs only in other nuclear reactions.
    If, for whatever reason, you need a neutron source in your hands this instant, a so called alpha source is the way to go. In this method, a radioisotope which emits high energy alpha particles (a few MeV) is mixed with beryllium. Polonium 210 is most likely the best option for this. The high energy alpha particle will fuse with the beryllium, emitting a neutron and leaving behing a carbon 12 atom. The neutron will have a few MeV of energy emission. Polonium can be purchased online, albeit at a steep price. http://www.imagesco.com/geiger/radioactive-sources.html#sources
    1 uCi, or one microcurie means just 37,000 alpha particles are emmitted per second, which means you would be very lucky to get over 10,000 neutrons per second per uCi. This obviously isn't the most effective method, but it is the one that requires the least effort and patience.
    I hope this helps!
  9. Jul 6, 2017 #8


    User Avatar
    2017 Award

    Staff: Mentor

    Bremsstrahlung can produce MeV photons, but it is not a very efficient process.

    A laser beam hitting an electron beam head on is another option. Backscattered photons receive a high energy. It is the method most sources of high-energetic photons use.

    There is an interesting proposal to use the LHC as an ultra bright source of high-energetic photons. It will be way too expensive to use it as a regular neutron source, but I think it is an interesting concept nevertheless. Besides scientific applications, it could be used for transmutation, e. g. to burn away nuclear waste.

    Currently the LHC accelerates protons and fully ionized lead atoms (Pb82+). When the particle physics programs run out (2035-2040), it could accelerate lead ions that keep a single electron (Pb81+) or similar ions. Shine visible/UV light against their direction of motion. In the lab frame these photons have several eV. In the frame of the lead ions they have a few keV to tens of keV - enough to excite them if the photon energy is right. They quickly return to the ground state and emit a photon with a few keV to tens of keV in their frame - which typically leads to tens to hundreds of MeV in the lab frame.
    The resulting beam would be millions of times brighter than existing MeV photon sources.

    Here is a presentation of the idea
  10. Jul 7, 2017 #9
    Oh, nice to see a discussion going on,

    well I already kinda knew that the only efficient way , atleast from a input energy/electricity bill viewpoint is to use natural fission as the "primary" and then either slow down those neutrons making them thermal or leaving them as fast ones depending on the requirement.
    But then again isn't this the way so called breeder reactors work? You essentially have a fission core surrounded by a fertile blanket and you let those fast neutrons hit the blanket and with time the blanket becomes filled with fissile material like Pu239/241 after the U238 has decayed into it. I assume molten salt reactors would have a benefit at this since this approach offers a steady chemical processing and "cleaning" from various elements of the breeder salt thereby increasing the neutron economy etc.

    Oh one question regarding this , i'm reading in the hyperphysics site that the French experimental fast breeder called the Superphoenix had it's core filled with quote" The reactor core consists of thousands of stainless steel tubes containing a mixture of uranium and plutonium oxides, about 15-20% fissionable plutonium-239. Surrounding the core is a region called the breeder blanket consisting of tubes filled only with uranium oxide. "
    So assuming that the Breeding fertile mostly U 238 blanket is the one on the outside not in the core does it then mean that they run the reactors primary fission core on MOX fuel?

    Also isn't the problem for fast breeders using sodium coolant that their separating structures aka the primary from the blanket and the reactor itself with respect to coolant are made from stainless steel of a special kind and even though its rather transparent to neutrons it does degrade from neutron structural damage much faster than the zirconium/steel and other alloys used in a typical thermal reactor core?

    As for fusion mentioned here as a neutron source, sure it sounds great as it has powerful neutrons but would it be fair to say that fusion as a neutron source as long as the fusion reaction itself does not yield power output greater than input is a rather wasteful and energy hungry process comparable to proton acceleration in particle accelerators?
    It would be great if we could somehow achieve break even fusion and have the ability to use the neutrons to breed a fissile fuel and then the reactor would become a two stage fusion/fission reactor in the exact opposite order as the thermonuclear weapons are two stage only in them fissile is the primary stage as here would fusion be the primary one. Although I'm not sure some of the more general fusion approaches could utilize such an approach because the tokamak relies on a blanket in which neutrons breed its own fusion fuel tritium from lithium so I guess it's a no.

    Oh one more while we are at it, the proposal about the LHC, you say mfb it could be used to burn away nuclear waste, my question is wouldn't that be a very energy demanding and expensive process in terms of waste burning because if I get the picture correctly one is essentially using a particle accelerator as the "primary" fission core that normally is in a fast breeder reactor and is used as the neutron source for the blanket? Only here instead of neutrons from natural fission which cost no extra money apart from the enrichment process are swapped for a beam of charged particles and a laser both of which are not very efficient in terms of power out vs electricity in correct?
  11. Jul 7, 2017 #10


    User Avatar
    2017 Award

    Staff: Mentor

    For the reactor questions, it is probably better if you open a thread in the nuclear engineering section.
    Both have advantages and disadvantages. The fusion approach is much more compact, for example, and produces less activation.

    It would be used for the most problematic waste only, energy demands are acceptable.
  12. Jul 7, 2017 #11
    But burning the waste with the lhc would be rather slow compared to a fast reactor or not? What are the particle densities compared between a fast core and a accelerator with the power in the range of Lhc?
  13. Jul 7, 2017 #12
    The fusion method is power hungry. Using the Gamow factor to calculate energy used per successful fusion, and treating that as an optimization problem, you get a minimum of 493.2 KeV per neutron in D-D fusion, and 295.6 KeV in D-T fusion. This translates to 2.195 KWh/10^20 neutrons fot D-D, or 1.316 KWh/10^20 neutrons with D-T. While the D-T is more efficient, each 10^20 neutrons will use 3.7 mg of tritium at this efficiency, or about $110.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted