Methane: The Ultimate Fuel?

[aftercolumbia]

Before (temporarily) going on the little tangent which gives this thread its title, I should clarify my intention to start a discussion of deep space maneuver stages along the lines of the Lockheed Martin studies which project possible efforts to make the Centaur stage operate for years.

Backgrounder 1: Hitchhiker's Guide To the Galaxy

According to Douglas Adams, there is a race of ancients who built an emormous, semi-sentient super computer specifically for the task of finding the Ultimate Answer.  When the computer had found the Ultimate Answer, it became extraordinarily obvious that the Ultimate Answer was absolutely useless without knowing the Ultimate Question.  Intergalactic historian Douglas Adams has some sort of disclaimer in his book saying that he made it all up, despite mounting evidence that this particular part of the story might not be so rediculous.  After Columbia has discovered two questions which have the Ultimate Answer, and speculates that one of them may be the Ultimate Question.

Ummm....am I forgetting something?  Yeah, oops, I guess I should share that the Ultimate Answer is "42".

Backgounder 2: After Columbia finds The Question candidates

The first question discovered with the Ultimate Answer was "How many sample modules should Judith carry?"  This discovery is already in the Mars Challenger report.

The second question discovered could well be "What is the Ultimate Fuel for near-term deep space propulsion?"

AFAL is an single dimensional isentropic engine simulator located at http://www.dunnspace.com/isp.htm.  It was created by the Air Force, provided to DunnSpace by a Mitchell Burnside Clapp, better known for his Pioneer Pathfinder booster stage, and certain general principles associated with it.  It has a catalog of approximately 323 different propellants you can chose to model a rocket engine with (including wood, coal and polytetraflouroethane (Teflon(R)), rocket propellants I do not plan to use anytime in the near future.)  Each propellant has a catalogue number.  For example, LOX is 37, while RP-1 is 63.

Studies about economical oxyhydrocarbon rocket propellant combinations to form alternatives to expensive hypergolics and oxyhydrogen, along with studies about how to passively maintain such propellants in deep space are constantly pointing to a single propellant combination as a possible "ultimate" propellant combination for current technology, which now includes V-shielded 60K radiators.  That propellant combination is oxymethane.  For AFAL simulations of this combination, two propellants are obviously selected:

LOX, with the number 37 and
CH4 (methane) with (you guessed it) 42.

Deep Space stage studies from Lockheed Martin:

http://www.lockheedmartin.com/data/assets/12382.pdf
http://www.lockheedmartin.com/data/assets/13349.pdf
http://www.lockheedmartin.com/data/assets/13350.pdf

On building modern space telescopes (which also like to be cold.)

http://www.rssd.esa.int/Herschel/Publ/1998/cryo98-fp-bc.pdf

[meiza]

This is a nice image from air liquide, shows the vapor pressure of methane at different temperatures:
http://encyclopedia.airliquide.com/images_encyclopedie/VaporPressureGraph/Methane_Vapor_Pressure.GIF

The forum will probably break the page url below:
http://encyclopedia.airliquide.com/Encyclopedia.asp?LanguageID=11&CountryID=19&Formula=&GasID=41&UNNumber=&EquivGasID=41&VolLiquideBox=&MasseLiquideBox=&VolGasBox=&MasseGasBox=&btnMSDS=0&MSDSLanguageBox=11&RD20=29&RD9=8&RD6=64&RD4=2&RD3=23&RD8=27&RD2=20&RD18=41&RD7=18&RD13=50&RD16=35&RD12=31&RD19=34&RD24=62&RD25=77&RD26=78&RD28=81&RD29=82&btnSubmitUnit=Click+to+change+the+values

Anyway, it shows what kind of temperatures can be achieved with methane with tank pressure constraints. At atmospheric pressure of 1 bar, 110 K. 10 bar you can sustain 140 K. So it doesn't rise very steeply.

Here's one for oxygen. 1 bar, 90 K. 10 bar, 120 Kelvin.
http://encyclopedia.airliquide.com/images_encyclopedie/VaporPressureGraph/Oxygen_Vapor_Pressure.GIF

Liquid methane smaller density (423 kg/m^3) than oxygen (1141 kg/m^3). If you use a mixture ratio of about 3, the tanks are roughly the same size.

Of course the performance is a lot crappier than with hydrogen.
You might already have browsed this methane engine list?
http://www.astronautix.com/props/loxlch4.htm
Seems the highest ones are in the 370 to 380 isp range. RD-0124 achieves 359 s with kerosene...

[Danderman]

Methane is a great fuel in theory, but it tends to generate soot in real world engines.

[aftercolumbia]

I looked at the "creosote problem" (I think creosote is a pretty word for an ugly thing) of oxymethane.  First, it is a lot milder than the same problem in oxykerosene.  I'm going to dig up some of my older Symtex GG numbers...when I did them I was aghast at the amount of solid carbon...then I remembered that Fastrac/MC-1 uses a fuel rich open cycle with oxykerosene.  I had a brief interest in using it, and did some research on it...so I happened to know it's gas generator mixture ratio was 0.3:1, and ran it in AFAL.

First, the Symtex numbers that made me gasp:
Mixture Ratio 0.55:1
Temperature: 1077.45K
Average Molecular mass: 14.850
H2: 5.338% (% = "G/100G" in AFAL)
H2O: 19.941%
CH4: 33.607%
CO2: 10.889%
C: 13.552% (the creosote)

The Fastrac numbers made me almost rush off to the washroom with a clean change:
Temperature: 1127.23K (differs from documentation's 889K, converted from 1600R)
Average Molecular mass: 20.378
H2: 4.584%
H2O: 10.372%
CH4: 18.693%
CO2: 5.601%
C: 43.584%

I thought if they could get that to run through a turbine...(not only did they, this is the engine that powered X-34)...

AFAL has not indicated solid carbon in main motor mixture ratios (2.7-2.9), but test videos from XCOR and InterOrbital seem to indicate that a small amount of solid carbon does actually form.  The biggest pain from this for the main motor (vs. oxyhydrogen and MON3/hydrazine) would be plume radiation to the payload...it would also be trivial compared to the plume radiation and case thermal soakback effects of solid motors (like the Delta II's Star 48B).  A lot of upper stages use hypergolic MON3/MMH, MON3/UDMH and both 25/75 mixture of UDMH and hydrazine, and a 50/50 mixture (better known as "Aerozine-50").  Those propellant combinations are not carbon free, and likely offer the best direct comparison for this particular issue.

[aftercolumbia]

meiza - 18/3/2007  6:52 PM

This is a nice image from air liquide, shows the vapor pressure of methane at different temperatures:
http://encyclopedia.airliquide.com/images_encyclopedie/VaporPressureGraph/Methane_Vapor_Pressure.GIF

The forum will probably break the page url below:
http://encyclopedia.airliquide.com/Encyclopedia.asp?LanguageID=11&CountryID=19&Formula=&GasID=41&UNNumber=&EquivGasID=41&VolLiquideBox=&MasseLiquideBox=&VolGasBox=&MasseGasBox=&btnMSDS=0&MSDSLanguageBox=11&RD20=29&RD9=8&RD6=64&RD4=2&RD3=23&RD8=27&RD2=20&RD18=41&RD7=18&RD13=50&RD16=35&RD12=31&RD19=34&RD24=62&RD25=77&RD26=78&RD28=81&RD29=82&btnSubmitUnit=Click+to+change+the+values

Anyway, it shows what kind of temperatures can be achieved with methane with tank pressure constraints. At atmospheric pressure of 1 bar, 110 K. 10 bar you can sustain 140 K. So it doesn't rise very steeply.

Here's one for oxygen. 1 bar, 90 K. 10 bar, 120 Kelvin.
http://encyclopedia.airliquide.com/images_encyclopedie/VaporPressureGraph/Oxygen_Vapor_Pressure.GIF

Liquid methane smaller density (423 kg/m^3) than oxygen (1141 kg/m^3). If you use a mixture ratio of about 3, the tanks are roughly the same size.

Of course the performance is a lot crappier than with hydrogen.
You might already have browsed this methane engine list?
http://www.astronautix.com/props/loxlch4.htm
Seems the highest ones are in the 370 to 380 isp range. RD-0124 achieves 359 s with kerosene...

Great pages, thanks for the links.  If I hadn't already known about them for months (have printouts with me), I'd be rather embarrassed.

You can also check out a lot of other stuff, such as hydrogen, helium, nitrogen, argon, xenon, and sulfur hexaflouride (a really boring cryogen used for system testing...where boring is a major plus.)

The equal volume mixture ratio is almost exactly 2.7 (Mars Challenger counts on this...saves having different tanks for each propellant.)  Symtex will use a common bulkhead that transfers heat freely between the propellants, so this ratio will decrease somewhat for Symtex, as the LOX has a lower boiling point.

[khallow]

Keep in mind that pure methane has low density for a fuel and a narrow temperature range over which it remains liquid. Also, it freezes at the boiling point of oxygen. One should also consider propane. It has a slightly lower ISP, but better fuel density (about 20% more dense I gather), is liquid at lox temperatures, and has a huge range over which it remains liquid (so it is easier to store in most places including Earth, Mars, and the Moon). Any time you have methane available, you can make propane.

[vda]

khallow - 20/3/2007  3:30 PM
One should also consider propane. It has a slightly lower ISP, but better fuel density (about 20% more dense I gather), is liquid at lox temperatures, and has a huge range over which it remains liquid (so it is easier to store in most places including Earth, Mars, and the Moon). Any time you have methane available, you can make propane.

Don't forget propane's relatives - cyclopropane and cyclobutane.

Cyclopropane (http://en.wikipedia.org/wiki/Cyclopropane) is interesting because it packs additional energy in it's highly twisted 'triangle' -> thus you get better Isp. However, it is a serious explosive hazard, so you can only use it as a small additive for propane - if they are miscible. I didn't find this info, but they are both non-polar and melting-boiling range is very close:

0.51 kg/L -187...-42 C - propane
0.72 kg/L  -91...12 C - cyclobutane
0.68 kg/L -128...-33 C - cyclopropane [nb: edited density here, was wrong]

If cyclopropane explosion hazard cannot be mitigated, is cyclobutane (http://en.wikipedia.org/wiki/Cyclobutane) good instead? It doesn't seem to be explosive, is commercially available, have better density than both propane and cyclopropane, and Isp *also* may be better than propane due to additional energy from strained bond.

[khallow]

vda, that sounds tricky. Both cyclopropane and cyclobutane have freezing points well above that of propane and the boiling point of oxygen. So that alone will limit how miscible they are (ie, too high concentrations will cause the additive to freeze out), if the propane is at lox temperatures.

[aftercolumbia]

khallow - 21/3/2007  7:30 AM

Keep in mind that pure methane has low density for a fuel and a narrow temperature range over which it remains liquid. Also, it freezes at the boiling point of oxygen. One should also consider propane. It has a slightly lower ISP, but better fuel density (about 20% more dense I gather), is liquid at lox temperatures, and has a huge range over which it remains liquid (so it is easier to store in most places including Earth, Mars, and the Moon). Any time you have methane available, you can make propane.

I noticed where it freezes.  Fortunately, I'm not contemplating tank pressures quite that low (i.e. <15psia), so it will stay liquid.  There is also ethane, the hydrocarbon between propane and methane.  When one is aiming for higher Isp (this is an upper stage), methane becomes a better choice than both ethane and propane.

The fuel that you are probably looking for is not cyclopropane (which is expensive, so I hear) but methylacetylene (BP 250K, MP 170K, density of 0.751 at 180K, main ingredient in MAPP gas).  LOX will freeze both methylacetylene and cyclopropane, so it looks like methane is still the right choice.

I'm hoping that a flexible propellant separation diaphragm can be developed that will allow liquid oxygen and methane to cohabit a single pressure vessel and use a single pressurization system injecting into the LOX volume.  As the stage operates, the diaphragm would move into the methane head, the methane staying 100% liquid via combination of LOX heat transfer and semi-active cooling (i.e. venting methane into a cooling jacket around the methane head, where it would expand and cool below the boiling point of the methane still in the tank.)  The methane side would not need a propellant management device, as the diaphragm is acting as a positive expulsion device for that side.  Also, a PMD in the methane side would keep the diaphragm from reaching the outlet, leading to high fuel residuals.  The LOX head does need a PMD because at the last ignition, it might be mostly gas.  It's the "last ignition" conditions that are rather critical.  With no LOX on the diaphragm (having been drawn away from it by the PMD), it becomes hard to transfer heat from the LOX to the methane side.  Because the methane side has no PMD, if any of it boils, you stand a very good chance of sucking bubbles (not to mention the lack of NPSHA at the engine inlet.)

One of the big advantages of the propellant separation diaphragm is that it makes it easy to vary the propellant load mixture ratio when loading the stage on the pad, without actually "offloading" the stage.  The short life variants are likely to have a higher mixture ratio than long life ones (which need to vent methane), but not the really long life ones (which don't need to vent anything).  It also makes engine upgrades easier, because when you increase the mixture ratio, you don't need to qualify a whole new tank and stage.

I should mention an inherent thought running through all this: the normal stage coast orientation is with the payload (and methane head) towards the Sun.  Alas, this could wreak havoc with remote sensing satellites that like to put their sensor platforms on top (i.e. Mars Global Surveyor, complete with 30deg sun angle avoidance zone.)  Hopefully, this is not a big deal.

[aftercolumbia]

Errata has given me a decent excuse to bump this one.  I'm intending to transfer heat from the methane side into the LOX side, not the other way around as I said in my previous post.

My Symtex study continues, with minimum energy payloads currently ranging from 671kg up to 1970kg.  Obviously, this fancy stuff is intended for higher energy missions.  I'm expecting it will be able to send some 246 to 400kg to Mars (246kg, actual number, 400kg...guess on an a large configuration run I haven't done yet.)  The most interesting mission this upper stage can be sent on is injection by an Atlas V 421 (or so) fully fuelled to Mars, where it inserts itself into a high energy orbit...and waits.

It would be waiting for a Mars sample return mission, one much larger than Mars Challenger, to launch its sample module into the stage's orbital plane.  While waiting, it might have been carefully aerobraking, primarily to control its line of apsides and overall inclination (i.e.: both inclination and RAAN) precession rates, so that it is in a decent orbit geometry for return to Earth.  After the sample module is launched, it lowers its orbit, docks with it, and propels it to an Earthbound trajectory.

[Christine]

Cyclopropane is right out, it costs more than $50/kg and has serious issues with pad handling.

Propylene though might also be interesting, particularly in a pressure fed system. Similar isp to cyclopropane, wont freeze at lox temperatures, and when chilled it has roughly the same density as kerosene. It's also used as feedstock in just about every industrial process known to man, so it's dirt cheap and readily available. It's no good with turbopumps though as it's vapor pressure is twice that of propane, and it's high combustion temps might make chamber cooling 'interesting'.

[yinzer]

I think propylene might have problems polymerizing in the combustion chamber cooling passages.

Also, having one propellant boil above the freezing point of another is not necessarily a show-stopper - oxygen freezes at a higher temperature than hydrogen boils, and yet the most successful LOX/LH2 stage uses a common bulkhead.

[E_ E_ H]

I'm producing a lot of methane at the moment, but then, I've been ill...  

[aftercolumbia]

Christine - 27/3/2007  7:28 PM
Propylene though might also be interesting, particularly in a pressure fed system. Similar isp to cyclopropane, wont freeze at lox temperatures, and when chilled it has roughly the same density as kerosene. It's also used as feedstock in just about every industrial process known to man, so it's dirt cheap and readily available. It's no good with turbopumps though as it's vapor pressure is twice that of propane, and it's high combustion temps might make chamber cooling 'interesting'.

More interesting is that it does have a propensity to polymerize, which is why it is so interesting as an industrial feedstock.  Assuming this isn't much of a problem in a rocket engine, the density advantage of pre-chilling it is relatively small for this type of stage (Dunn's study prechills the LOX as well, this one would not.)  The vapor pressure curve is such that it is unlikely to be of any use as a duration coolant, which is one of the reasons I like methane.

Also, all of these fuels contain more carbon than methane, which could be a pain in fuel-rich open-cycle turbopump engines.

[aftercolumbia]

yinzer - 27/3/2007  8:40 PM

I think propylene might have problems polymerizing in the combustion chamber cooling passages.

Also, having one propellant boil above the freezing point of another is not necessarily a show-stopper - oxygen freezes at a higher temperature than hydrogen boils, and yet the most successful LOX/LH2 stage uses a common bulkhead.

Lockheed Martin documents also say that the LOX tank is single largest heat source for the Centaur's LH2 supply!  (That counts the sun.)  Of course, such a characteristic (boiling/freezing conflict) is more likely to be a problem in a propellant separation diaphragm stage.  Of course, the onus is simply to keep the tank pressure high enough that the LOX vapor/liquid equilibrium temperature is above the freezing point of methane.  That is pretty easy.  After the first maneuver (which is probably LEO ascent completion), even that can be relaxed as the LOX PMD is going to pull the liquid oxygen away from the PSD.  With vapor on the PSD, the heat from the methane side will not transfer through very quickly, especially in microgravity.

[publiusr]

I think RD-0120 (Energiya core block engine) could run on methane according to Wade's site.

[aftercolumbia]

I doubt it.  It's a fuel-rich staged+expander cycle engine based on...well...inspired by the SSME.  Burning methane in it would re-introduce the creosote problems and require a total redesign of the injector and associated manifolds to accomodate the much higher volumetric mixture ratio.

Also, the RD-0120 is far too large for Symtex.  One of them puts out nearly as much thrust as the largest Symtex variant does at lift-off, and about 140 times as much as the oxymethane stage.

[publiusr]

Look up RD-0120M-CH

Channel Wall might help with any coke problems.

[aftercolumbia]

I'm am bumping this thread because of the test thread elsewhere in this forum.  This thread should be able to answer some of the questions without having to repost a bunch of stuff.

[khallow]

BTW, earlier in this thread there was some discussion of the "creosote problem" when one runs a fuel-rich burn with methane (or kerosene). I gather this is the build up of carbon and other carbon-based compounds on any surface exposed to the burning fuel or exhaust. How serious is this problem? And what activities (eg, engine restart) are affected by this?

For example, it seems to me that a straight burn of a few minutes (say from a first stage boosting out of the atmosphere) wouldn't matter. OTOH, using a methane engine for lifting off from Mars and occasional maneuvering back to Earth over many months could have serious problems with carbon buildup. Could a burst of oxidizer clear the buildup? Ie, run the engine for a few seconds with an oxidizer rich mixture?

Where does the problem manifest? Fuel injectors? Inside of the reaction chamber and exhaust outlet? Outside of the vehicle near the engine?