Main question: Is it possible to achieve net power generation based on linear proton accelerator and U-238 target?

In the proposed reactor design there is a proton beam with energy ~10 GeV, and on U-238 target where these protons should generate cascade neutron generation & U-238 fission with high-energy neutrons.

The problem though is that there is some disagreement on energy looses and neutron multiplicity.

My estimations and understanding is the following:

  1. 60% of incident proton energy goes into heat.
  2. For high energy protons (>1GeV) there are significant loses on pi-0 decay, so using proton energy above 1-1.5GeV is inefficient. Is there other significant ways of loosing energy?
  3. Estimated number of neutrons generated is 120-250 per incoming 10 GeV proton on Uranium target.
  4. Only fraction of such neutrons have energy high enough to support fission of U-238
  5. If only 25% of these neutrons would cause 1 fission, we'll only have ~180/4*0.2=9 GeV of fission + comparable amount of energy in neutrons and high-energy gamma (from pi-0 decay)

Does that seems to be correct? Is there any papers investigating possibility of spallation-based net power generation on U-238/Thorium reactors? (i.e. without classical fissile materials like U-235, U-233 or Plutonium)

Also, based on published specs of SINQ-2 spallation neutron generator, I calculated it's neutron generation efficiency to be 0,024% of proton beam power. Even if we replace spallation target with Uranium one, it's efficiency would be 0,1662% at best. Obviously, with such efficiency we cannot support net power generation. Does that mean that this idea is not workable?

  • $\begingroup$ Nice question. My gut expectation is that this won't work simply because accelerators are woefully inefficient machines, but I don't have enough figures on hand to argue it. $\endgroup$ Commented Jun 11, 2014 at 22:12

1 Answer 1


This is a nice idea that has been studied thoroughly, and continues to being studied. You are talking about a type of sub-critical reactors referred to as Accelerator Driven Systems (ADS). Here is a nice review Review:Basics of accelerator driven subcritical reactors, where references to previous studies are given. For modern advances you might like to check the iTHEC website.

But just so you know before going into the literature, Thorium has been found to be the most promising fuel for this idea for several reasons, and its burning cycle goes through $^{233}U$.

  • $\begingroup$ Well, this is not ADS. The difference is that we are not waiting for accumulation of fissile isotopes, but rather burn U-238 directly using relativistic neutrons. $\endgroup$ Commented Jul 20, 2014 at 19:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.