Would this lead to a reactor that produced more energy than you put in?


This has actually been proposed! Project PACER at Los Alamos explored the possibility of detonating hydrogen bombs in underground cavities (e.g. in salt domes) to produce "geothermal" energy. Exploding a 2 kton device once every 20 minutes in a steel lined underground cavity using molten salt droplets to cushion the effects of the explosive would produce about 1000 MWe of power. This is a fusion reactor since most energy comes from the fusion reaction, started by a nuclear detonation.

  • 3
    $\begingroup$ The fact that someone proposes an idea does not make it in any way sensible. How many active projects to build such a device are under way ? A 2kT explosion every twenty minutes ? Doesn't that sound just the tiniest bit crackpot to you ? $\endgroup$ – StephenG Aug 25 '19 at 11:12
  • 1
    $\begingroup$ @StephenG - Peer reviewed publications. Check the calculations, they made sense to me. I would expect it to be very impractical, but the question was never if using bombs to power reactors was a good solution to energy production, just if it was possible. $\endgroup$ – Anders Sandberg Aug 25 '19 at 15:05
  • $\begingroup$ No argument about the source you link to (good find, BTW), however peer reviewed, while a better gauge for quality material, does not make an idea sensible. The OP asked about power production and that, to me, means practical. Strictly speaking the OP asked if more energy is produced than you put in and, again strictly speaking, the answer to that is no. It only makes sense as a question if you're considering the practicality of it all (IMO). $\endgroup$ – StephenG Aug 25 '19 at 16:14
  • $\begingroup$ @StephenG - More energy DOES come out. Strictly speaking, the answer is absolutely yes. What are you talking about? $\endgroup$ – Maury Markowitz Oct 10 '19 at 20:49
  • $\begingroup$ And as to "practical", well you only know that if you try. Some chunk of LLNL was working on this under Plowshares to the point they carried out actual UGX's to test it. I don't know why you would call this crackpot. A good portion of LLNL also worked on mirror fusion, it didn't work in the end either, are you saying Post is a crackpot too? $\endgroup$ – Maury Markowitz Oct 10 '19 at 20:51

I think this confuses what is hard about getting fusion reactors to work.

I assume you are thinking of using a thermonuclear device to start a fusion reactor rather than to use one or more as a fusion reactor. In other words I assume you are not thinking of harnessing the energy released from one or more such devices as in the various mad schemes people involved with nuclear weapons thought up to harness them for peaceful power generation. That approach is addressed in another answer.

The problem with getting a fusion reactor to work is not getting fusion started (anyone can do that): it is keeping the stuff you are fusing stable and contained for long enough that you can get more power out than you put in. There are various approaches to doing this of which we know of only one really successful one: that approach – gravitational confinement – unfortunately doesn't scale down to scales we can use directly.

Given that the problem is not starting fusion but containing it and keeping it stable, it seems unlikely that detonating a thermonuclear weapon inside your fusion reactor is going to help!


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