I'm working on a fusion rocket concept for a hard science fiction setting. I'm envisioning laser catalyzed Proton-Boron reactors (low waste heat, aneutronic, 100% of their output being He4 Alpha particles) based on HB11 or TAE's designs, but with two major drawbacks:

1) They must convert almost all of their particle output into electricity in order to break even, leaving only a small amount that can be used for thrust

2) The lasers themselves need to be replaced or refurbished after a few hours of use.

So when all's said and done, the engine performs more like a nuclear thermal rocket than a torch ship or an Epstein drive. Stylistically and tech wise, 'fusion but it's not that good' would fit my setting like a glove.

Plausible concept or dumb idea?

  • 1
    $\begingroup$ Welcome to Worldbuilding.SE! We're glad you could join us! When you have a moment, please click here to learn more about our culture and take our tour. Did you mean "hours of use?" That's an economically poor period of time. Even if you used them for two hours, drifted for days on end, then used them to slow down... it's like having to overhaul the engines after every trip. Anyway, I recommend replacing the science-fiction tag with the reality-check tag. $\endgroup$
    – JBH
    Mar 15, 2019 at 23:05
  • 1
    $\begingroup$ Since I had to look up the links for HB11 and TAE I just added the links to the question without asking you. Hope you do not mind. Also hope those are the correct links. $\endgroup$ Mar 16, 2019 at 3:14
  • $\begingroup$ @JBH Thank you, I'll make sure to read them thoroughly. Several hours before overhaul isn't really as bad as it sounds. If I recall correctly, 10 hours is the lifespan for an SNRE engine, and most burns for cislunar space last only a few minutes. Most proposed Mars missions using NTRs have burn times of about half an hour. $\endgroup$
    – Jake
    Mar 16, 2019 at 3:23
  • $\begingroup$ Atomic rockets for hydrogen boron : projectrho.com/public_html/rocket/… Rostoker (ie TAE) is mentioned but the HB11 variant (which like the question uses lasers) is not. Probably because its performance is unknown. $\endgroup$ Mar 16, 2019 at 3:28
  • $\begingroup$ @VilleNiemi I actually sent HB11 an email a while back, and they said that their design could be used for propulsion in space but not for STO. Low thrust, but I'm envisioning LH2 afterburners being standard. $\endgroup$
    – Jake
    Mar 16, 2019 at 3:40

2 Answers 2


This sounds plausible indeed.

Any engine can be abstracted as producing two things: thrust and waste heat. As an intermediate step, it can produce internal power, but said power will either be used for helping thrust, directly or indirectly, or will end up as waste heat due to other inefficiencies. (I am ignoring big third-party energy drains like, say, giant laser weapons.)

Here, most of the output is used for feeding the laser/particle beam/plasma that is keeping the reaction going. But this power isn't going anywhere: it is either used to move particles faster, or to heat things up (due to unavoidable inefficiencies, as waste heat). So you can simply describe your engine as being, say, 50% or 90% efficient. Which means that the remaining 50% or 10% remaining power needs to be dealt with. With a spacecraft that cannot use atmosphere, water or ground to evacuate heat (like an airplane, ship or ground powerplant would), this means giant red-glowing radiators. Which is a good thing! Radiators look badass, they convey the feeling of power well, they are a good visual cue for what is happening (cold radiators mean the powerplant is stopped, for example), they are a great source of plot and complication, and hard-SF fans will talk about how realistic your work is. In fact, the biggest mistake of The Expanse (both book and show) is probably to not have used them.

Now, propulsion is done by throwing particles as fast as possible in the other direction. The faster you can throw the particles, the longer the same mass of fuel will last you. Fusion drives are great for that: with their enormous power, they can emit particles at an enormous speed, and thus are very efficient. Unfortunately, they also have very low thrust, because a fusion drive will comparatively emit a low mass of particles per second - think better ion drives. Contrast with chemical engines that have a terrible efficiency (you need a whole Saturn V to send a tiny capsule to the moon and back), but have enormous thrust.

Note that, with the lack of radiators (implying an efficiency of 99.99% at least) this is why the Epstein drive is implausible (even if not physically impossible).

You can actually increase thrust with the same power by using more propellant, but as such each particle will be emitted at a lower speed (same power divided between more particles). So you have more thrust but less efficiency. You can even use your engine in maximum powerplant mode, and use the energy to heat inert propellant (, thus having the thrust of a chemical engine for a slightly better efficiency. In fact, if you are in atmosphere, you can even use air and turn it into a jet engine. Do remember that even proton-boron fusion is somewhat radioactive and not quite aneutronic due to secondary reactions.

So for interplanetary travel, you want maximum efficiency, with longer, gentler burns. For manoeuvre down gravity wells, you want better thrust because short, harder burns are more important. Do play at Children of a Dead Earth if you want to experiment with this.

Your reactor can even be designed with this inefficiency directly: if it is too complicated to have a 100% fusion rate, maybe it uses much more hydrogen and/or boron than is ultimately fused, and the remainder is heated by the reaction and used for thrust along with the fusion results. So while theoretical max specific impulse is very high, maybe your engine has actually a lower one.

As for changing laser heads every few hours, this wouldn't have too much effect on the efficiency of the engine as such, being equivalent to a slightly less efficient one (the mass of the spare laser heads). However, it is a good way to make it feel like a high-tech, high-maintenance device, and opens things up for plot points and complications - again, a good thing. After all, we don't know the exact design compromises that go to a spacecraft fusion laser, so there is no reason for it not to feel believable.

And of course, Atomic Rocket is the website to go for when working on hard-SF.


Lets call it - whack a mole fusion. You give up on the idea of control on the process. And just whack at spikes of plasma coming near the edge of containment.

How this is done?

You use a set of lightning-rod lasers (https://ec.europa.eu/digital-single-market/en/news/fet-open-laser-lightning-rod) to 3d print a coil shape on the plasma of the emerging spike.

Once that spiral of fog is printed onto the inferno, you send current through, creating a counter turmoil, that compresses the plasma back into the containment region. It needs great sensors, fast reaction times and its definatly wastefull and slow, with output constantly fluctuating. In addition you can not control this whole thing and it could run hot in regions with constant magnetic fluctuation.

In addition its error prone- meaning a split second to late and the coil printed might already be distorted and shove the plasma into a unexpected direction.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .