If it's possible, could you use "hybrid" ion engines, maybe VASIMR, running on a mix of hydrogen, argon, and oxygen to exceed the velocities produced by any other ion engine? Could buckyballs also be used?

Explanation for the hydrogen; https://space.stackexchange.com/a/34503/29415 VASIMR

The VASIMR does not require a grid.

  • 1
    $\begingroup$ +1 The idea here being that exothermic reactions from some of the gasses might help heat the plasma, or just contribute to the exhaust velocity? $\endgroup$
    – uhoh
    Mar 1, 2019 at 3:26
  • 1
    $\begingroup$ Both, actually. $\endgroup$ Mar 1, 2019 at 13:35

3 Answers 3


If you're going to burn the LH with LOX, you'll melt the ion engine components that are in the exhaust path, e.g. the grids. An ion engine needs these to function.

enter image description here

Throttling down until this is not an issue leaves you with low-pressure, inefficient combustion of the LH (if it can combust at all).

After expending the LH/LOX, you're left with two heavy tanks that will make your spacecraft heavier.

You're better off separating them into 2 engines:

  1. a LH/LOX stage to provide an initial $\Delta$V, which gets separated after burnout.
  2. a normal ion engine.

Ion engines can work for years because they use storable propellants. LH/LOX boils off, so you either have to burn it quickly, or use (heavier) insulated tanks.

  • $\begingroup$ Or control the flow of the LOXY and LHY so that they make up less of the volume of the exhaust, and last longer than with a large flow. $\endgroup$ Mar 1, 2019 at 13:37
  • $\begingroup$ The VASIMR doesn't require a grid $\endgroup$ Mar 1, 2019 at 18:36
  • $\begingroup$ Therefore, with an electromagnetic force keeping the plasma away from the walls, there would be no tarnishing. $\endgroup$ Mar 1, 2019 at 20:19
  • $\begingroup$ The forces involved in an ion engine are much smaller than the ones in a chemical engine. Electromagnets strong enough to contain the LH/LOX combustion would make the ion engine far too heavy to be practical. Again, you're better off separating them. $\endgroup$
    – Hobbes
    Mar 2, 2019 at 8:12
  • $\begingroup$ VASIMR has antennas in the exhaust path. Same problem. $\endgroup$
    – Hobbes
    Mar 2, 2019 at 8:13

In addition to Hobbes specific answer for this specific case in general for any engine system the first stage of acceleration reaction mass from stationary is easy and the last element is hardest, if only because by that point your exhaust mass is moving so fast it spends so little time in the engine. If you have an efficient acceleration method boosting reaction mass from from 1kms to 1.1kms it also be efficient going from 0.1 to 0.2kms as well, and overall performance is all about optimising the final exhaust velocity (such as by not contaminating it with corrosive compounds or heating things that need to be electrically conductive).

There is a slight corollary here with starting from zero, since actually getting your exhaust stream moving at all may actually involve a low effeciency 'booster' that gets your reaction mass to right temperature, charge condition and pressure for optimal operation through a given electrical engine but the design aim will be to minimse the energy/mass being used in the low efficiency stage and to maximise velocity change in the high efficiency stage.

So a future high performance electrical engine might in fact use some form of chemical 'fuel' to achieve optimal conditions for the actual used reaction mass, but it would be to the minimum needed to optimise the electrical engine performance rather than as a thrust source in its own right.


Ideally, you want the propellant to go straight out the back as fast as possible. Any sideways motion, i.e. from heat, is wasted kinetic energy.

That's one way to think about what a conventional rocket engine's nozzle does: By expanding the exhaust, it cools it and converts that (sideways) thermal energy to desirable velocity toward the back.

So if you burn your propellant before running it through the ion engine, you'll lose performance unless you find a place to put an expansion nozzle before the ion engine. And since it's an expansion nozzle, you'll need a Really Big Bore on the ion acceleration channel: That adds mass.

You'd be better off using the burning fuel in a conventional engine, dropping it away, and then starting an ion engine. Each could then be optimized.


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.