For the Apollo missions in the 1960s, NASA used a combination of kerosene and liquid oxygen to get the rockets out of Earth's atmosphere.

Why did they do that? If you look at the statistics, the second stage and beyond for the Apollo rocket had a combination liquid oxygen and liquid hydrogen while only the first stage had kerosene.

Does the mixture of oxygen and hydrogen provide a sort of advantage in space than Earth?

I'm asking this question mainly because I'm working on a sort of Cities In Space project and am wondering if there is a better or more efficient way of escaping Earth's atmosphere.


Hydrogen-oxygen rocket engines are the most efficient chemical rockets that are reasonably safe and practical to use, by propellant mass. The metric normally used for this comparison is called specific impulse or Isp, (typically given in units of seconds for historical reasons) and for the Saturn V's upper stages it's about 421 seconds. For comparison, kerosene-oxygen specific impulse usually runs around 300 seconds (for first-stage engines of that era).

For upper stages, mass efficiency is incredibly important, because the lower stages have to lift that mass. Thus hydrogen-oxygen is an extremely common choice for that role.

For the first stage, however, mass efficiency is much less important than cost. Kerosene-oxygen engines produce much more thrust per dollar for a number of reasons: kerosene is far denser than hydrogen, so you build a physically much smaller stage, reducing assembly and transport costs; hydrogen plumbing is much trickier; kerosene's energy density means the engine is physically smaller and thus easier to build, transport, install, etc.; liquid hydrogen must be kept much colder than liquid oxygen, and so forth.

(The Q/As linked by @uhoh in comments illustrate the implications of these tradeoffs for Saturn V.)

The same pressures that made kerosene attractive for the Saturn V first stage apply even more to solid rocket engines -- lower Isp still, but much more compact, simple, and cost-effective, which is why you see them used as boosters in many modern launchers.

  • 1
    $\begingroup$ I'm curious... what are the more efficient rockets that aren't safe or practical? $\endgroup$ – paj28 Oct 11 '16 at 9:53
  • 5
    $\begingroup$ @paj28 Tripropellant lithium-hydrogen-fluorine is the highest isp possible, but it's one of the least practical as well. $\endgroup$ – Agent_L Oct 11 '16 at 10:06
  • 2
    $\begingroup$ Yup. Basically anything with Fluorine is high-performance bad news -- it is tempting because it's a highly energetic oxidizer, but it's terribly dangerous to work with because it's a highly energetic oxidizer. $\endgroup$ – Russell Borogove Oct 11 '16 at 10:20
  • $\begingroup$ @Agent_L Just out of curiosity, what kind of Isp does one get out of that mix? $\endgroup$ – a CVn Oct 11 '16 at 13:28
  • 1
    $\begingroup$ @michaelkjorling Li-H-F specific impulse is around 515s (5050 m/s Ve) in vacuum per Huzel & Huang by way of Wikipedia. en.m.wikipedia.org/wiki/… $\endgroup$ – Russell Borogove Oct 11 '16 at 15:32

My personal favorite source of Saturn V information, SP-4206 "Stages to Saturn," weighs in on the RP-1 choices.

From Chapter 7:

When the contract to build the biggest stage of the Saturn V, the S-IC first stage, was awarded to Boeing on 15 December 1961, general outlines of the first-stage booster were already fairly well delineated. The main configuration of the S-IC had already been established by MSFC, including the decision to use RP-1, as opposed to the LH2 fuel used in the upper stages. Although LH2 promised greater power, some quick figuring indicated that it would not work for the first stage booster.

Liquid hydrogen was only one half as dense as kerosene. This density ratio indicated that, for the necessary propellant, an LH2 tank design would require a far larger tank volume than required for RP-1. The size would create unacceptable penalties in tank weight and aerodynamic design. So, RP-1 became the fuel. In addition, because both the fuel and oxidant were relatively dense, engineers chose a separate, rather than integral, container configuration with a common bulkhead. The leading issue prior to the contract awards related to the number of engines the first stage would mount.

That chapter goes into a lot of detail about the design of the first stage tanks, which are enormous as-is. I think it's fair to say that even larger LH2 tanks would've compounded some of the construction problems that were had (though they could've been overcome).

Chapter 4 is more about the engines, and implies that technology readiness is a factor:

NASA's contract award to Rocketdyne in 1959, calling for an engine with a thrust of 6.7 million newtons (1.5 million pounds), was a significant jump beyond anything else in operation at the time. Executives within the space program looked on the big engine as a calculated gamble to overtake the Russians and realize American hopes for manned lunar missions. It seemed within the realm of possibility too, by using engine design concepts already proven in lower thrusts and by relying on conventional liquid oxygen and RP-1 propellants.

It's common conservatism in aerospace engineering to take incremental steps forward, so for the newly developed high-thrust F-1 engines for the first stage, they stuck with otherwise-proven hydrocarbon fuels.

I'm not immediately finding a clean source for this last claim, so it might deserve to be edited out, but I believe there's also a specific thrust advantage owing to the density of the fuel, and so for a first stage with a relatively short burn time it can be more efficient for the overall system to pay the specific impulse penalty to get the bird off the ground and out of dense atmosphere with the additional thrust, then drop the stage and switch propellants to something higher-impulse. Similar arguments apply for the strap-on solid boosters commonly used with several launch systems today.

  • $\begingroup$ LH is actually far less than half as dense as kerosene; I suspect the comparison was supposed to be (LH+LOX) versus (kerosene+LOX). $\endgroup$ – Russell Borogove Oct 11 '16 at 5:58
  • $\begingroup$ Concur on the relative densities--should I change the last paragraph to read "specific thrust advantage to RP1"? Or is there another place that I made a mistake? $\endgroup$ – Erin Anne Oct 11 '16 at 6:19
  • $\begingroup$ No, your conclusion is even more correct than the quoted material suggests. $\endgroup$ – Russell Borogove Oct 11 '16 at 8:43
  • $\begingroup$ Oh, gotcha. That is a weird thing in the source--haven't run the numbers on it but my guess is even (LH+LOX) is well less than half as dense as (RP-1 + LOX), especially if the Kerolox is run fuel-rich. $\endgroup$ – Erin Anne Oct 12 '16 at 2:29
  • $\begingroup$ Yeah, if I remember right it's like 3.5:1 depending on the mix ratios. $\endgroup$ – Russell Borogove Oct 12 '16 at 3:42

Its probably got to do with burn rate, the second fuel burns faster to give a quicker push. Once you near the burning velocity of the first fuel it is less effective at accelerating you. A rocket is accelerated by the explosion hitting the roof of the chamber. Summary: its like first and second gear in your car.


Some of the information contained in this post requires additional references. Please edit to add citations to reliable sources that support the assertions made here. Unsourced material may be disputed or deleted.

  • $\begingroup$ Neither an internal combustion engine nor a rocket engine works by explosions, controlled or otherwise. $\endgroup$ – a CVn Oct 11 '16 at 13:14
  • $\begingroup$ Explosion or not, exhaust velocity is a factor, and He has a higher exhaust velocity (and Isp) than kerosene. So far so good. But exhaust velocity is not the reason kerosene is chosen for the first stage. $\endgroup$ – Hobbes Oct 11 '16 at 13:51
  • $\begingroup$ He is not used as fuel, but H is used. Burning He with oxygen is not possible, He is only used for pressurization. $\endgroup$ – Uwe Oct 12 '16 at 13:53

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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