IIRC, tripropellant rockets can deliver superior performance compared to bipropellant rockets (at the cost of added complexity). Now, I know that there is the Li/H/F combo, but it isn't used because of how dangerous it is. The thing is, there were another few tested:

  • Be/H2/O2 (obviously not used because of beryllium)
  • Al/RP-1/O2 (doesn't sound too dangerous to me)
  • Al/H2/O2 (doesn't sound too dangerous to me)

and a few more. What happened to these?

Edit: Following up on a few thoughts provided in the comments, why isn't more research being done on less toxic combos when the reward could be great?

  • 1
    $\begingroup$ Also, source: ntrs.nasa.gov/citations/19860018652 $\endgroup$ – xilpex Feb 20 at 3:54
  • $\begingroup$ Related: space.stackexchange.com/questions/5939/… $\endgroup$ – Russell Borogove Feb 20 at 5:08
  • 3
    $\begingroup$ Skimming that doc, it looks like the aluminum triprops only provide very modest increase in theoretical specific impulse for H2/O2 and none for RP-1/O2. Rocket engine development is expensive and rocket designers are therefore conservative; I'd guess no one has decided it's worthwhile to spend the money to develop such an engine for a 1% increase in Isp. $\endgroup$ – Russell Borogove Feb 20 at 5:17
  • 2
    $\begingroup$ Injectors are very difficult, tankage is too massive the increased efficiency, you need 50% more turbopumps, etc... $\endgroup$ – Anton Hengst Feb 20 at 5:35
  • 1
    $\begingroup$ @xilpex it isn't obvious that there's much to gain, though. Modern boring rocketry is quite economical and reliable for launch, and ion engines have the high Isp-in-vacuum thing sewn up... is there likely to be much return on investment? $\endgroup$ – Starfish Prime Feb 20 at 22:36

Al = aluminum is a solid metal, RP1, O2 and H2 are stored as liquids within the tanks of a rocket.

If you know a method to store a liquid preparation of aluminum that may be pumped into a rocket tank and after ignition pumped into a combustion chamber, we all here would like to read about it.

Molten hot alumium is not useful in a rocket tank, it needs heating to stay liquid. A combination of three different temperature levels, the very cold oxygen, the petrol at room temperature and the very hot liquid alumium within the same rocket would require a lot of heavy insulation and a heat source to keep the aluminum liquid.

A suspension of powdered alumium in rocket petrol would separate into powder at the bottom of the tank and petrol above it.

Powdered aluminum is used for rockets, but only as a part of solid fuel, like the boosters of the Space Shuttle.


Simply because the rewards aren't all that great, and haven't merited the added complexity. The potential improvements of tri-propellant mixes are limited by fairly simple chemistry (especially once you eliminate the options with toxicity or cost issues), and to get them you're adding another tank, another pump, more plumbing, all of which adds dry mass that eats away at any potential advantage.

Costs are more related to operational complexity than to propellant performance. For example, just compare SpaceX's RP1-burning Falcon 9 to the "higher performance" but much more expensive Delta IV. SpaceX's in-development Starship launch system continues this approach: it burns methane due to its availability and ease of handling, and uses a stainless steel structure rather than some ultra-light composite.

Particularly relevant to this question, SpaceX intends for Starship to use an autogenous pressurization system that uses vaporized propellant to pressurize the tanks instead of helium, specifically to reduce the number of fluids that the vehicle and ground support equipment must handle. They are willing to add design complexity (the evaporators and such required for autogenous pressurization) to reduce operational complexity by eliminating one fluid, one which isn't even a propellant.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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