The ISP of a propellant is important when mass is the limiting factor; however, that is not exactly the case when launching a rocket. For the first part of the flight, the density of the propellant is also important.

Higher density → more propellant in the same tank → higher mass ratio → more $\Delta v$

A better metric for propellant performance for the first km/s is therefore density impulse. It is easy to notice that the extremely low density of liquid hydrogen (70.8 kg/m³) makes methane, ammonia and water outperform it as a NTR propellant in the early stages of the flight.

In the search for an even better density impulse, I tried to find high density metals with a low enough boiling point. Mercury and zinc turned out as possible candidates. For some reason, I am a bit afraid of glowing-hot, vaporized, radioactive mercury rain, so I choose instant galvanization instead. Zinc has an excellent density impulse. At an estimated 130s of ISP, the 7140kg/m³ density implies a density impulse 15 times better than hydrogen, and over twice that of water. Furthermore, as thrust is inversely proportional to exhaust velocity at the same reactor power, this is also giving the rocket a better thrust to weight ratio.

But, zinc is not a fluid, so I wonder, is it possible to have a mix of zinc powder and water or ammonia as a propellant? Is it possible to pump such a mix through the engine and prevent the reactor from clogging? Do you know of any research related to this?

Worked example of density impulse:

Sample rocket:

Dry mass: 10,000kg
Tank volume: 100m³

Sample propellants:


Exhaust velocity: 9000 m/s Density: 70.8 kg/m³

$$\Delta v=9000m/s \cdot ln\left(\frac{10,000kg + 100m^3 \cdot 70.8 kg/m^3}{10,000kg}\right)$$

$$=4818 m/s$$


Exhaust velocity: 4000 m/s Density: 1000 kg/m³

$$\Delta v=4000m/s \cdot ln\left(\frac{10,000kg + 100m^3 \cdot 1000 kg/m^3}{10,000kg}\right)$$

$$=9592 m/s$$

Feel free to leave out any discussion on the feasibility of using a NTRs, or potential benefits from a high density impulse from your answer.

What I really want to know is if it is possible to pump metal powder through an engine when mixed with water, ammonia or methane. If it is possible, how high concentrations of metal powder I can have, and still be able to pump it?

  • 4
    $\begingroup$ Do you know any research related to this? Yes, John D. Clarke's Ignition! An Informal History of Liquid Rocket Propellants. Page 177 (chapter High Density and the Higher Foolishness) mentions mercury and page 9 uses Diethylzinc to get hypergolic engine start. Not as main fuel though. $\endgroup$
    – Hennes
    Commented Jan 10, 2016 at 15:06
  • 3
    $\begingroup$ Three possible complications: neutronics, localized cooling, radiation safety constraints on going with a hot NTR on the first stage. $\endgroup$ Commented Jan 10, 2016 at 20:44
  • 2
    $\begingroup$ You can circumvent the 'zinc is a solid' problem by having a coolant loop from the reactor to the zinc tank, melting the zinc before you use it. $\endgroup$
    – Hobbes
    Commented Jan 26, 2016 at 9:47
  • 8
    $\begingroup$ What's wrong with "glowing-hot, vaporized, radioactive mercury rain"? HA... $\endgroup$ Commented Jan 26, 2016 at 13:53
  • 2
    $\begingroup$ @SF. Using a NTR is kind of like getting rid off the oxidizer... $\endgroup$ Commented Jan 27, 2016 at 11:11

1 Answer 1


I suggest you look into the chemistry of clay slip and slurries. For example, porcelain slip is very fluid and can pour through distribution channels in molds to form fine features, then sets up as a weak solid after losing only a small amount of water to the porous mold. The percentage of water is quite low, with the liquid properties coming from carefully controlling the properties of the porcelain clay and a small percentage of surfactants and lubricity compounds.

So in a word, yes, you can have a room-temperature liquid slurry comprised of a particulate solid, lubricity compounds, and a pretty small percentage of water or other carrier liquid. HOWEVER you face problems with settling and clogging. Since the water content is so low, any place you drive a density gradient, such as in a pipe elbow, will cause solids to settle out forming a clog.

If you can consider liquid metals rather than a particular slurry, there are some Lead-Bismuth and Tin-Bismuth alloys that melt at or below 170°C. Zinc casting alloys melt around 720°F.

I thought you were crazy pushing the density impulse idea until I remembered that Aerojet demonstrated boosting 1st stage thrust in a (simulated) NTR using LOX injection at the nozzle inlet. Great thrust/weight for the first portion of the flight, then switching to a high-ISP, low-thrust/weight mode once the heavy mass-additive is expended, with very little parasitic tankage weight remaining.


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.