Fuel Types

[Kerbalsaurus]

As we all know, rockets are powered by many different fuel types. From the iconic LH₂+LO to highly experimental iodine and electrical fuels, rocket fuel can greatly vary. But how many are there exactly? There has to be more than just those.

[DDE]

A brief perusal of Ignition! An Informal History of Liquid Rocket Propellants should get you up to speed. A comprehensive list doesn't exist because it would involve digging through several nations' MILSPEC storeable propellant specifications.

[K^2]

It's hard to come up with something you can't use as a rocket fuel. Water, just water, is a pretty decent monopropellant when superheated. At room temperature the heat of vaporization is 2.25MJ/kg. This gets a little worse at high temperatures, which you generally want for better thrust, but it's still a lot of energy. For reference, kerolox is about 7.54MJ/kg, because you need a lot of oxidizer. Consequently, steam rockets are a thing, and would probably be more common if the tank wasn't so heavy, severely eating into the useful dV.

So when you can just boil some water and get yourself a useful rocket fuel, is it a surprise that exotic combinations are practically innumerable?

[DDE]

32 minutes ago, K^2 said:

So when you can just boil some water and get yourself a useful rocket fuel, is it a surprise that exotic combinations are practically innumerable?

Absent all other context, it would be. A layman may expect a "dominant" choice to arise (e.g. hydrolox) whereas in practice the myriad of applications, each with their own developmental idiosyncrasies, ensure countless niches.

[Kerbalsaurus]

3 hours ago, DDE said:

Absent all other context, it would be. A layman may expect a "dominant" choice to arise (e.g. hydrolox) whereas in practice the myriad of applications, each with their own developmental idiosyncrasies, ensure countless niches.

Yes. These are words I understand.

[Abel Military Services]

Does anybody mentioned metallic hydrogen? I've been playing KSPI-E for a while, so I know it. 

Edited December 28, 2022 by Abel Military Services

[magnemoe]

On 12/23/2022 at 11:49 PM, DDE said:

Absent all other context, it would be. A layman may expect a "dominant" choice to arise (e.g. hydrolox) whereas in practice the myriad of applications, each with their own developmental idiosyncrasies, ensure countless niches.

Some cubesat engine plan was to use water, they split it into gas hydrogen and oxygen and burn this for trust  Benefit is high ISP and simple, obviously pulsed as you need to split water using the solar panels before use. 

For most uses its an limited number of fuels in common use. Hydrogen, RP1, methane, hypergolic and solid fuel for launches, not sure if its huge variation is solid fuel for large rockets. 
Hypergolic is slowly going away it makes sense for missiles but orbital rockets don't need to stand ready to launch and the fuel is dangerous. 

In space you have hypergolic again also monopropelant, ion engines is also very popular because very high ISP. Its work on safer green monopropelants, not so much for environment but for easier handling.
Then you have the weirder stuff like my first example. Cubesats are mostly secondary payload and on budget so they can rarely use dangerous fuel who can endanger primary mission. 

[K^2]

17 hours ago, magnemoe said:

Some cubesat engine plan was to use water, they split it into gas hydrogen and oxygen and burn this for trust  Benefit is high ISP and simple, obviously pulsed as you need to split water using the solar panels before use.

ISP is still significantly better for ions. Even your cheap grid RF iodine thruster, typical for cubesats, can do 2,000s+, while the absolute best you can do with hydrogen/oxygen bursts is something like 500s, and even that's probably not achievable in a cubesat. But the low thrust of these simple ions is a pretty big limitation in the kind of maneuvers you can do. If you want to actually do orbital transfers, being able to do short burns near apsides gives you a lot more freedom over the weeks-long ion burns. So going the electrolysis route is a very neat workaround. If I wanted to get a cubesat from LEO to LLO, I'd likely be prepared to sacrifice some of that ISP for the burst capability. Even if I'm unlikely to get more than 5m/s per burst.

[magnemoe]

19 hours ago, K^2 said:

ISP is still significantly better for ions. Even your cheap grid RF iodine thruster, typical for cubesats, can do 2,000s+, while the absolute best you can do with hydrogen/oxygen bursts is something like 500s, and even that's probably not achievable in a cubesat. But the low thrust of these simple ions is a pretty big limitation in the kind of maneuvers you can do. If you want to actually do orbital transfers, being able to do short burns near apsides gives you a lot more freedom over the weeks-long ion burns. So going the electrolysis route is a very neat workaround. If I wanted to get a cubesat from LEO to LLO, I'd likely be prepared to sacrifice some of that ISP for the burst capability. Even if I'm unlikely to get more than 5m/s per burst.

Yes, should have specified high ISP for non ion,  and as you say it give an short burst of power who is nice for some settings. 

 

[sevenperforce]

If you want a nice look at various propellant combinations (even some pretty exotic ones), you can peep the table on this page. It has all of the data from the Joint Army-Navy-NASA-Air Force Interagency Propulsion Committee reports on propellant combos, and provides useful information like exhaust velocity, mixture ratio, chamber temperature, and propellant bulk density, both at sea level and in vacuum.

If you want examples of actual propellant types used in actual rocket engines, this wikipedia page is a fantastic resource. I personally created the "Power Cycle" column for this page, actually, back in July of 2021. It has power cycle, propellant combo, specific impulse, and more.

[jimmymcgoochie]

If it burns, it can (*probably) be used as rocket fuel. If we’re only talking about liquid-fuelled chemical rockets here then there’s a long and varied list of propellants that have been used, or at least tested, from the common to the bizarre to the outright terrifying. These include (not an exhaustive list!):

• Refined kerosene, aka jet fuel, energy dense and pretty easy to get hold of in large quantities, usually burnt with liquid oxygen although in some cases (cough Britain cough) hydrogen peroxide was used instead;
• Liquid hydrogen, lightweight and very efficient but also very un-dense, extremely cold and prone to boiling away and escaping through the walls of your fuel tanks, usually burnt with liquid oxygen although hydrogen-fluorine rockets were proposed at least once as fluorine is a better oxidiser than oxygen;
• Liquid methane, offering a half way point between kerosene’s density and hydrogen’s efficiency and cryogenic properties, a few real rockets are going to use methane as their fuel but so far no rocket has made it to Earth orbit using methane, typically used with liquid oxygen;
• Hydrazine, a compound of hydrogen and nitrogen, able to be stored for long periods and used as either a monopropellant for RCS or small rockets or as a bipropellant, hypergolic and spontaneously combusts when it contacts a suitable oxidiser, usually nitrogen oxides;
• Mono- and dimethylhydrazine, derivatives of hydrazine with extra methyl groups added, widely used in hypergolic rockets such as ballistic missiles as they can be stored for long periods in missile silos and launched at short notice, also used for RCS, upper stages and almost all US crewed spacecraft, typically used with nitrogen oxides although one Soviet engine combined UDMH (unsymmetrical dimethylhydrazine) with liquid oxygen, see also Aerozine-50, a proprietary blend of hydrazine and UDMH used by rocket engines made by Aerojet;
• Pentaborane, absolutely ghastly stuff that’s as toxic as nerve gas but provides a higher ISP than other hypergolic fuels, the Soviet RD-270 engine was converted to use pentaborane (as the RD-270M) with nitrogen tetroxide but fortunately never flew, other boranes are similarly energetic but similarly toxic too;
• Chlorine fluorides, WHY!!?!??! Toxic, corrosive, will burn with almost everything including sand, asbestos, water, fuel tanks, engine components, humans… DO NOT USE!
• Ethanol, used in some of the earliest liquid fuel rockets including the V-2 missile and some versions of the Redstone missile used for suborbital Mercury capsule launches, poor ISP compared to kerosene and prone to mysteriously going missing from the storage tanks, usually used with liquid oxygen or nitric acid.
• Aniline, or aminobenzene, used in very early US sounding rockets with furfuryl alcohol and nitric acid;
• Cavea-B, a monopropellant researched in the US to replace hydrazine but ultimately abandoned due to safety concerns, decomposes violently in a self-sustaining reaction once started with a dab of UDMH hence the safety concerns of the reaction racing up the pipes and into the fuel tanks;
• Liquid ammonia, considered for a Soviet upper stage engine combining it with fluorine but this never made it to reality;
• Liquid lithium, just don’t ask how you’d keep lithium molten in a rocket with liquid fluorine in close proximity and gaseous hydrogen nearby too and just look at that crazy high ISP! Tested with an ISP of well over 500 seconds, well above even hydrolox, but impractical and lithium has since become very valuable for batteries.

There are other, more niche propellants out there and rare combinations such as a kerosene-hydrogen-oxygen tripropellant rocket, and that’s not even mentioning solid rockets, nuclear thermal rockets or ion/plasma/electric propulsion systems or the most famous rocket propellant of them all: Coke and Mentos.

[StrandedonEarth]

8 hours ago, jimmymcgoochie said:

Ethanol, used in some of the earliest liquid fuel rockets including the V-2 missile and some versions of the Redstone missile used for suborbital Mercury capsule launches, poor ISP compared to kerosene and prone to mysteriously going missing from the storage tanks, usually used with liquid oxygen, nitric acid, or ice cubes

FTFY

Nice writeup!

[Superluminal Gremlin]

On 12/24/2022 at 7:28 AM, DDE said:

Ignition! An Informal History of Liquid Rocket Propellants

I have a paper copy of this book. An absolute banger. Someone does die however (Some metal, titanium i think was in some acid (for tests), exploded and killed the guy next to it.)

It also talks about tripropellant motors, namely Li-F-H and Be-H-O.  The first one had a ridiculous impulses of 542 seconds!. Sadly, the exhause had, well, Fluorine in it.  And slurries. Those are weird.

Compound A (ClF₅)

CTF - Chlorine Tetrafluoride

Flox 30 and 70 - Fluorine and oxygen

IRFNA

IWFNA

LOX (duh)

MMH

RFNA

UDMH

WFNA

plus at least 50 other special little chemicals which never really worked.

The book eventually states "Lithium Hydride and a Hydrocarbon binder with Chlorine Tetrafluoride and Perchloryl Fluoride is possibly one of the better combinations". Sounds expensive, and dangerous, but oh my, the ISP would be worth it 

It also talks right at the end about free radicals. But the only people ever known to have success trapping them is the FBI...

[kerbiloid]

[Abel Military Services]

Did anyone mentioned nuclear powered engines? 

Liquid hydrogen - used in NTRs, and sometimes fusion engines for better thrust

Nuclear salt water(Water with Uranium salt or Plutonium salt) basically weapons grade fissile materials in water, it is very dangerous, be careful when you store it! When the nuclear salt water goes into the engine, the fission will happen and you will get a very high thrust! 

Fusion fuel(Usually Deuterium or Helium-3) I think I doesn't need to explain that. 

[farmerben]

This is new and cool!

[sevenperforce]

On 1/3/2023 at 1:54 AM, Abel Military Services said:

Did anyone mentioned nuclear powered engines? 

Liquid hydrogen - used in NTRs, and sometimes fusion engines for better thrust

Nuclear engines are of course terrific fun. It bears pointing out, however, that the sort of nuclear thermal engines we can currently build do not get as hot as many of the chemical engines we can build. They are more efficient not because they are more energetic than chemical engines, but because they supply their own heat and so you can use a lightweight propellant like pure liquid hydrogen without needing to bring along any heavy oxidizer.

[Terwin]

On 12/27/2022 at 8:45 PM, Abel Military Services said:

Does anybody mentioned metallic hydrogen? I've been playing KSPI-E for a while, so I know it. 

Unfortunately, the only form of metallic hydrogen we have evidence for thus far would require very high pressure containment, thus the tanks could very well weigh more than an order of magnitude more than the fuel contained therein.  This would make it useless as a rocket fuel.

If metastable metallic hydrogen exists(which is what they have in KSPI-E, I believe), then that would be a wonderful rocket fuel, but unfortunately, the only paper that suggests metastable metallic hydrogen is even possible(as far as I am aware) has been debunked(to the best of my knowledge), and there is not currently any strong hope that metallic hydrogen can exist at pressures where it would be useful for rocket fuel.

 

On the other hand, you can combine just about anything with ClF3 and get a strong exothermic reaction, too bad it is so hard to handle and has very toxic combustion products.

Edited January 6, 2023 by Terwin

[Beamer]

Gummy bears...

 

[jimmymcgoochie]

 

54 minutes ago, Beamer said:

Gummy bears...

 

My chemistry teacher once stuck a jelly baby into a test tube with molten potassium permanganate at the bottom, triggering a reaction that briefly caused the jelly baby to get stuck in the tube until gas pressure fired it out the end, on fire, with significant velocity, whereupon it immediately burned a hole through the curtains.

More of a cannon than a rocket really, but a fun science lesson nonetheless.

[DDE]

2 hours ago, Terwin said:

On the other hand, you can combine just about anything with ClF3 and get a strong exothermic reaction, too bad it is so hard to handle and has very toxic combustion products.

AFAIK ClF5 is superior all-around...

Except for that bizarre case where somebody "cooked" a souffle by bubbling ClF3 through liquid F2. For some reason, pentafluoride just didn't work.

[AckSed]

I recommend digging into these archives of the Usenet group sci.space.tech. There are some real gems gleaned from moderately famous rocket scientists (including people involved in the DC-X and Roton) and general all-round smart geeks. It's not at all up to date, with few of the posts dated after 2000, but I learned things I'd never have found out otherwise:

Propargyl alcohol (H-C-C-CH2-OH) is a room-temperature liquid fuel that is denser than kerosene and produces slightly better performance than it when burned with liquid oxygen, and is even better with high-test hydrogen peroxide, a room-temperature liquid oxidiser. The latter combination is nothing you'd want to splash around, but it's not half as bad as hydrazine and its pals.

Propane (C3H8)/liquid oxygen has one or two nice tricks that make it a bit better than methane. Like methane, it can be used for combustion chamber or nozzle cooling, converting it into a warm gas to feed into the injectors, improving mixing. It also boils off into the atmosphere once testing is finished, making reusability easier. Unlike methane, once it's chilled down to LOX temperatures it has similar density to kerosene, and its melting point overlaps with the boiling point of oxygen, allowing it to share a common bulkhead with the LOX tank, saving weight.

Carbon Monoxide (CO)/LOX could be a cheap, simple rocket fuel to produce on Mars or in the atmosphere of Venus, as while performance is poor, it needs no hydrogen and can be made directly from their CO2 atmospheres.

Ethyl alcohol/ozone (O3) was tried, but the ozone was not only more oxidising than fluorine, it was too damn unstable. Diluting O3 in liquid oxygen made it marginally safer, but it had a higher boiling point than O2, so once testing was finished evaporation would concentrate it out into high-concentration O3, and then explode. Violently.

Furfuryl alcohol/white fuming nitric acid (HNO3) has a bit of history as the first practical hypergolic liquid bipropellant. Performance wasn't that good, though, and making the nitric acid storable involved adding nitrogen tetroxide. It's still available in bulk and is cheap, though.

Polyamide plastic/Nitrous oxide (N2O) is the basis of one of the few hybrid solid/liquid rocket engines actually flying (or flown) - on Virgin Galactic's SpaceShipTwo. It's cheap and relatively safe, and allows throttleability by restricting the nitrous as needed.

Propane/N2O, or NOP for short, was test-fired enough to gain a patent, because N2O can be catalytically decomposed like peroxide, but is self-pressurising, providing pressurisation for its own tank and for the propane tank, stores as a dense liquid but injects as a gas and is less finicky than peroxide. It can even be used as a monopropellant for attitude control and ullage thrusters. Propane's advantages are already known, but the combination is low-risk, easy to handle and both are available in bulk.

Kerosene/peroxide is one of the might-have-been-greats. Both liquid at room temp., both safe to moderately safe, and with clever design the peroxide not only cools the engine, the heat (and/or a catalyst) decomposes it into oxygen and steam hot enough that it automatically lights the kerosene. It launched a satellite. It's even clean-burning. But US rocketry is really skittish about peroxide, and chemical suppliers won't sell it in the higher concentrations needed.

Edited February 5, 2023 by AckSed

[AckSed]

Coming back to this, I realised I had not covered tri-propellant mixtures, and those are interesting, not least because some combos pushed the envelope of materials science and even sanity in trying to make the damn things work. In no particular order:

Fluorine/oxygen/any, AKA FLOX has been talked about from time to time (and upthread), despite the difficulties of working with fluorine. Apart from the modest increase in specific impulse, when used with RP-1 or another hydrocarbon, it is hypergolic, simplifying engine starts, yet has enough oxygen to react sufficiently with the carbon.

Oxygen/hydrogen/kerosene is actually a compelling combination, except they are not always used together together. Rather, in the two engines I know of, kerolox is used in the ascent for the high thrust, then it switches over to hydrolox for greater specific impulse once the rocket has ascended high enough that that matters. The interesting part is where this switchover is applied. The MAKS spaceplane would have been powered by the RD-701, using separate turbopumps in each combustion chamber for H2/O2/kero and shutting down the kero once it was depleted. The other example was Aerojet's invention of thrust-augmented nozzles, injecting the kerosene downstream of the combustion chamber (and producing an interesting shock pattern). The cool thing about this was it let the engine use an over-expanded nozzle but without the instabilities of flow separation because of the higher pressure in the nozzle. Not so cool was the tradeoff in specific impulse, but it didn't matter because of the extra thrust at the beginning of the ascent. Hilariously, a TAN might also be used in a pseudo-tripropellant hydrogen/oxygen/oxygen setup: injecting extra oxygen into the nozzle of a hydrolox engine to burn stochiometric, and thus increase thrust by throwing out more water.

Fluorine/lithium/hydrogen is infamous. Take one of the strongest oxidisers known, combined with hot, liquid lithium that is hypergolic with air and corrodes any gasket material known, and then throw in cryogenic hydrogen. But oh, the shiny, shiny specific impulse. It had a measured vacuum impulse of 542s with a long nozzle, with the ferocious heat of the Li-F reaction leaving the hydrogen acting like the propellant in a nuclear-thermal engine. But oh, the exhaust products. This would only ever be considered as a vacuum engine, but the handling difficulties (and the difficulty of packing a liquid metal at the same temp. as a deep fat fryer into the same rocket as two cryogenic liquids) would be immensely annoying.

Aluminium/oxygen/helium is a new one on me, but the more I read, the more attractive it becomes, at least as a lunar-ISRU fuel. (At the time, water on the Moon was not a sure thing. The other alternatives, sulphur/O2, magnesium/O2 and phosphorous/O2 give me a bit of pause.) As we know, a rocket throws out reaction mass in the form of molecules. It's simplified, but the more heat these molecules have, the more energy they have. The heavier molecules in the exhaust, like CO, H2O and CO2, give more thrust, while the lighter molecules like H, H2 or helium have more velocity for the same amount of energy. That's what this does: take the heat of combustion of Al/O2 and flow 5-10% of the lightest inert gas through the combustion chamber to reduce the molecular weight and increase specific impulse. Theoretical calculations predicted an impulse of 373s from Al powder carried in 10% LHe, but acknowledged that a) actual experiments would be needed b) helium would need to be brought from Earth.

Hydrogen/beryllium/oxygen was only mentioned in passing, but they give the impression that it's theoretically good, but a horror practically. Beryllium is one of those metals that's fine as a block, but really quite toxic in powder form, which is what it would need given its high melting point. It burns extremely hot, which is great for a rocket. However, it creates beryllium oxide, which has an extremely high melting point. As soon as the engine cuts out it then condenses out into a powder coating the inside of the nozzle, injectors, combustion chamber and throat... and it's toxic and carcinogenic as well.

[mikegarrison]

I really think ethanol is the ideal rocket fuel. If it doesn't get you to orbit, at least it can get you to where you don't care about not getting to orbit.

Edited February 16, 2023 by mikegarrison

[sevenperforce]

7 hours ago, AckSed said:

Fluorine/lithium/hydrogen is infamous. Take one of the strongest oxidisers known, combined with hot, liquid lithium that is hypergolic with air and corrodes any gasket material known, and then throw in cryogenic hydrogen. But oh, the shiny, shiny specific impulse. It had a measured vacuum impulse of 542s with a long nozzle, with the ferocious heat of the Li-F reaction leaving the hydrogen acting like the propellant in a nuclear-thermal engine. But oh, the exhaust products. This would only ever be considered as a vacuum engine, but the handling difficulties (and the difficulty of packing a liquid metal at the same temp. as a deep fat fryer into the same rocket as two cryogenic liquids) would be immensely annoying.

Hydrogen/beryllium/oxygen was only mentioned in passing, but they give the impression that it's theoretically good, but a horror practically. Beryllium is one of those metals that's fine as a block, but really quite toxic in powder form, which is what it would need given its high melting point. It burns extremely hot, which is great for a rocket. However, it creates beryllium oxide, which has an extremely high melting point. As soon as the engine cuts out it then condenses out into a powder coating the inside of the nozzle, injectors, combustion chamber and throat... and it's toxic and carcinogenic as well.

You will enjoy this short story very much.

A relevant excerpt:

Spoiler

He paused for a mouthful of lime margarita. “Take the space shuttle,” he said wistfully. “With just two tweaks, we could have put a hundred tons into its payload bay!”

“Two tweaks?” I asked doubtfully. A hundred-ton payload (in a vehicle already massing close to a hundred tons) would have put the shuttle in the same bracket as the Saturn V.

“Yes.” He smiled sourly. “They could have stretched it, given it a bigger thermal protection system as well—the Columbia disaster wouldn’t have happened. But they rejected my proposal. The first part, to upgrade the SRBs, would have been trivially easy! Although the alternate oxidizer for the space shuttle main engines would have presented certain handling difficulties, that much is true . . .”

“Tell him about the SRBs first,” Jim suggested. He leaned forward expectantly; at a guess, he’d heard this before.

“All right. First, the solid rocket boosters. Regular SRBs run on a mixture of ammonium perchlorate—the oxidizer—and finely powdered aluminum, suspended in a rubbery polymer that holds everything together and provides additional reaction mass. When they ignite you get aluminum oxide and ammonium chloride and lots of energy. But it’s not really enough! We could make them about twenty percent more efficient if we just replaced the aluminum with powdered beryllium. It’s a lighter atom and the redox reaction is more energetic—”

“Hang on!” I stared at him. “Beryllium is really poisonous. Wouldn’t that—”

Leonard shook his head. “Nonsense.” A small smile. “You see, then there was my second proposal. If you replace the oxidizer in the space shuttle main engines with liquid fluorine, you could also get an extra twenty percent out of them. And I know what you’re going to say next: wouldn’t that give rise to an exhaust plume of extremely hot hydrofluoric acid? You’re absolutely right: it would! But hydrofluoric acid reacts with beryllium oxide to give you beryllium fluoride—which is almost inert in comparison—and hydrochloric acid, which is neither here nor there.” A shadow crossed his face. “It’s totally safe, compared to some of the other projects I’ve worked on. But NASA took one look at the environmental impact statement and, and . . .” His shoulders began to shake; whether with laughter or tears, I couldn’t tell.

 

7 hours ago, AckSed said:

Oxygen/hydrogen/kerosene is actually a compelling combination, except they are not always used together together. Rather, in the two engines I know of, kerolox is used in the ascent for the high thrust, then it switches over to hydrolox for greater specific impulse once the rocket has ascended high enough that that matters. The interesting part is where this switchover is applied. The MAKS spaceplane would have been powered by the RD-701, using separate turbopumps in each combustion chamber for H2/O2/kero and shutting down the kero once it was depleted.

A fuel-switching tripropellant engine has always been one of my favorite proposals for a 1.5-stage-to-orbit reusable launcher.