Several internet articles claim that the high Isp of the Soviet RD-180 engine (as used in the Atlas III and V launch vehicles) was (at least partially) due to the fact the RD-180 ran oxidizer-rich, unlike its fuel-rich American contemporaries. These articles provided no references to support the assertion of higher efficiency for oxygen-rich combustion.

EDIT: According to Wikipedia, the combustion chambers of the RD-180 ran fuel-rich. It was their LOX pre-burner which ran oxygen-rich.

EDIT:https://ntrs.nasa.gov/search.jsp?R=19820002372 deals with relative advantages of oxidizer- and fuel-rich pre-burners but not combustion chamber mix ratios.

My understanding is that maximum temperature (and combustion efficiency) is attained with a stoichiometric mix of fuel and oxidizer. However, these high temperatures exceed the service temperature of available engine materials. This relationship is illustrated by data for methane in gas turbines:

enter image description here


As a result, chemical rocket combustion chambers are designed to run fuel- or oxidizer-rich for durability reasons. Either fuel- or oxidizer-rich conditions will lower combustion temperature, but fuel-rich has the added advantage of lower exhaust gas molecular weight and therefore higher exhaust velocity ( Isp).

As a result, I would expect rocket designers to favor fuel-rich conditions to maximize Isp efficiency.

Fuel-rich engines have coking problems. Oxidizer-rich engines are very hard on metal parts. Both these are disadvantages but do not affect theoretical efficiency. Which strategy is objectively more efficient (e.g.: Isp or TWR)? Are there any successful oxidizer-rich engines?

  • 2
    $\begingroup$ The RD-180 does not run oxygen-rich. It runs at a O:F ratio of 2.72:1, stoichiometric would be something like 3.4:1 (going by the C[n]H[1.953n] formula from braeunig.us). The "OR" in ORSC refers specifically to the preburners. $\endgroup$ Commented Oct 10, 2023 at 16:19
  • 3
    $\begingroup$ Does this answer your question? What are the advantages of Ox-rich staged combustion? $\endgroup$ Commented Oct 10, 2023 at 16:47
  • 1
    $\begingroup$ maximum temperature (and combustion efficiency) is attained with a stoichiometric mix of fuel and oxidizer Maximum specific impulse is generally achieved with a fuel-rich mix, because it leaves simpler molecules in the exhaust which hide less energy in interatomic bonds than complex molecules. $\endgroup$ Commented Oct 10, 2023 at 16:49
  • 1
    $\begingroup$ Thanks for the heads-up! The link in that answer has been fixed. $\endgroup$ Commented Oct 10, 2023 at 17:11
  • 2
    $\begingroup$ @DarthPseudonym you are exactly right. Hot oxygen makes pretty much everything want to burn. The Soviets worked some metallurgical magic to get their engines to work. $\endgroup$ Commented Oct 10, 2023 at 22:00

1 Answer 1


Regardless of whether oxidizer-rich engines would be more efficient, the RD-180's efficiency is not due to that, as the RD-180 runs fuel-rich. According to braeunig.us, for each carbon atom in RP-1 there are 1.953 hydrogen atoms, so a stoichiometric ratio would be something like 3.4:1. The RD-180 runs at a O:F ratio of 2.72:1, definitely on the fuel-rich side of stoichiometric.

What distinguishes the RD-180 and similar Russian engines is not that they run oxidizer rich (as they don't), but that they are staged combustion engines. This does several things, but the most significant is that the working fluid used to pump propellant is injected into the combustion chamber to be combusted further and used as reaction mass, so much higher pumping powers can be used, allowing higher chamber pressures than gas generator engines. Their preburners run oxidizer rich, which allows them to do staged combustion with carbon-rich fuels that would otherwise cause coking that would clog the turbine passages and engine injectors.

At the time these engines were developed, the only staged combustion engines developed in the US used fuel-rich preburners with hydrogen fuel which didn't produce coking, and the only kerosene-burning engines used gas-generator or other cycles that were better able to handle the coking issues. Being able to use staged combustion gave kerolox engines like the RD-180 a major performance edge over other kerolox engines, while retaining the thrust and ease of handling advantages of kerosene fuel.

  • $\begingroup$ Technically "coking" happens in the liquid fuel and "sooting" happens in the combustion products, but your answer covers it well. space.stackexchange.com/a/59099/6944 apps.dtic.mil/sti/citations/ADA410860 archive.org/download/engine-cycles-2/EngineCycles-2.pdf ntrs.nasa.gov/citations/19820002372 $\endgroup$ Commented Oct 11, 2023 at 23:47
  • $\begingroup$ @OrganicMarble that's usually the case, but the mixture ratio in a FRSC preburner is far more fuel-rich than that in gas-generator engines. Most of the fuel wouldn't really participate in combustion beyond being heated by it. In a kerolox FRSC, it would probably actually be coking (thermal decomposition and polymerization) instead of sooting (incomplete combustion resulting in carbon deposits). That said, I'm not aware of such a thing being any more than a hypothetical possibility, or maybe a (probably short-lived) laboratory experiment. $\endgroup$ Commented Oct 12, 2023 at 3:08
  • $\begingroup$ Is a gas generator cycle limited to the achievable chamber pressure, at least in a practical sense to the 1000psi-class engines (F-1, RD-107)? i.e. do "high chamber pressure" and "staged combustion" almost necessarily go together? $\endgroup$ Commented Oct 12, 2023 at 12:12
  • 1
    $\begingroup$ @DanielChisholm See en.wikipedia.org/wiki/Comparison_of_orbital_rocket_engines. In a gas generator cycle, increasing the chamber pressure requires increasing the amount of propellant going to the gas generator + turbine used to pump the propellant. That propellant then gets dumped overboard without producing significant thrust. There is an optimum point where increasing the chamber pressure further reduces effective overall performance. Staged combustion engines don't have this issue. $\endgroup$ Commented Oct 12, 2023 at 13:05

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.