Refueling Mars-Bound Starships with Lunar LOX - Calculations & Discussion Thread

[TheRadicalModerate]

Okay, but can "thin walled" tanks survive the mass driver/air gun/sling launcher/whatever?

Well, LOX is paramagnetic...

Update:  This was sorta-kinda a joke, but I didn't realize just how much of a joke until I went to find the permeability of LOX.  All I found was the susceptibility, which is 1.52E-3.  That means that μ/μ₀ = 1.00152.

[Nevyn72]

Wot?

[Twark_Main]

For now, a placeholder destination for transferring post from NSF Purgatory.

See title. Don't Panic.

[spacenut]

So a Mars bound Starship leaving Earth with a full load of liquid methane makes a pit stop at LL1 or LL2 to pick up lunar lox, and then on to Mars. 

Or, a fuel depot at a Lunar station that has liquid methane brought from Earth and lox brought from the moon.  Then a Starship stops by to fuel up and then on to Mars. 

Would either save time to Mars, or save earth launched tanker Starships?  Tankers re-entering Earth's atmosphere from LEO has less heat and stress than a tanker taking methane to the moon for fueling there and returning to earth. 

Seems like option 1 above might be better and have fewer tankers to just get the Starship to cis-lunar for top off of lox. 

[geekesq]

So a Mars bound Starship leaving Earth with a full load of liquid methane makes a pit stop at LL1 or LL2 to pick up lunar lox, and then on to Mars. 

Or, a fuel depot at a Lunar station that has liquid methane brought from Earth and lox brought from the moon.  Then a Starship stops by to fuel up and then on to Mars. 

Would either save time to Mars, or save earth launched tanker Starships?  Tankers re-entering Earth's atmosphere from LEO has less heat and stress than a tanker taking methane to the moon for fueling there and returning to earth. 

Seems like option 1 above might be better and have fewer tankers to just get the Starship to cis-lunar for top off of lox.

Or you could refuel at a Lunar station using lunar-manufactured LOX and lunar-manufactured methane, given that scientists think it is highly likely there is frozen carbon dioxide (as well as frozen water) on the moon:

"Carbon dioxide cold traps on the moon are confirmed for the first time"
https://phys.org/news/2021-11-carbon-dioxide-cold-moon.html

CO₂+2H₂O+energy = CH₄+2O₂

[kevindbaker2863]

since we are talking  100's if not thousands of ships to mars, it would be optimal to identify an orbit for the fuel depot to maximize fuel for Mars-bound flights.   so whether the methane comes from the earth or the moon what is the orbit that supports mars but is not too costly to get to from where the fuel is coming from.  might wind up having depots in different places depending on where fuel comes from?

[LMT]

...scientists think it is highly likely there is frozen carbon dioxide (as well as frozen water) on the moon:

"Carbon dioxide cold traps on the moon are confirmed for the first time"

Caveat:  the paper just maps cold spots.  Some might lack CO2 ice, especially if, say, some impact heating event redistributed CO2.

But lunar carbon sources did exist, especially prior to 3 Gya.  And Cabeus LCROSS impact did actually release some carbon, including CO2, (Table 2), consistent with mapped Cabeus "carbon dioxide cold traps", Fig. 2.

Refs.

Colaprete, A., Schultz, P., Heldmann, J., Wooden, D., Shirley, M., Ennico, K., Hermalyn, B., Marshall, W., Ricco, A., Elphic, R.C. and Goldstein, D., 2010. Detection of water in the LCROSS ejecta plume. science, 330(6003), pp.463-468.

Schorghofer, N., Williams, J.P., Martinez‐Camacho, J., Paige, D.A. and Siegler, M.A., 2021. Carbon dioxide cold traps on the Moon. Geophysical Research Letters, p.e2021GL095533.

[meekGee]

...scientists think it is highly likely there is frozen carbon dioxide (as well as frozen water) on the moon:

"Carbon dioxide cold traps on the moon are confirmed for the first time"

Caveat:  the paper just maps cold spots.  Some might lack CO2 ice, especially if, say, some impact heating event redistributed CO2.

But lunar carbon sources did exist, especially prior to 3 Gya.  And Cabeus LCROSS impact did actually release some carbon, including CO2, (Table 2), consistent with mapped Cabeus "carbon dioxide cold traps", Fig. 2.

Refs.

Colaprete, A., Schultz, P., Heldmann, J., Wooden, D., Shirley, M., Ennico, K., Hermalyn, B., Marshall, W., Ricco, A., Elphic, R.C. and Goldstein, D., 2010. Detection of water in the LCROSS ejecta plume. science, 330(6003), pp.463-468.

Schorghofer, N., Williams, J.P., Martinez‐Camacho, J., Paige, D.A. and Siegler, M.A., 2021. Carbon dioxide cold traps on the Moon. Geophysical Research Letters, p.e2021GL095533.

Am I reading it wrong or are the (uncertainties) in the compounds containing carbon about equal to the full scale values?

[LMT]

...scientists think it is highly likely there is frozen carbon dioxide (as well as frozen water) on the moon:

"Carbon dioxide cold traps on the moon are confirmed for the first time"

Caveat:  the paper just maps cold spots.  Some might lack CO2 ice, especially if, say, some impact heating event redistributed CO2.

But lunar carbon sources did exist, especially prior to 3 Gya.  And Cabeus LCROSS impact did actually release some carbon, including CO2, (Table 2), consistent with mapped Cabeus "carbon dioxide cold traps", Fig. 2.

Refs.

Colaprete, A., Schultz, P., Heldmann, J., Wooden, D., Shirley, M., Ennico, K., Hermalyn, B., Marshall, W., Ricco, A., Elphic, R.C. and Goldstein, D., 2010. Detection of water in the LCROSS ejecta plume. science, 330(6003), pp.463-468.

Schorghofer, N., Williams, J.P., Martinez‐Camacho, J., Paige, D.A. and Siegler, M.A., 2021. Carbon dioxide cold traps on the Moon. Geophysical Research Letters, p.e2021GL095533.

Am I reading it wrong or are the (uncertainties) in the compounds containing carbon about equal to the full scale values?

As they explained, yes.  See also a subsequent analysis of LCROSS volatile release, with notable addition of CO, in Berezhnoy et al. 2012.

--

An alternate prospecting tool for lunar carbon, the Carbon Favorability Index (CFI) of Cannon 2021, highlights several craters.

Image:  Cannon 2021 Fig 3.  Lunar polar CFI, with carbon-prospect hotspots.

--

Often overlooked: 

LCROSS released appreciable NH3 at Cabeus.  That might be significant. 

Rationale:  Space Force et al. have strong interest in ASCENT propellant, due to its high density, easy long-term storage in depots, etc.  Formula:  [NH3OH][NO3].  Note requirement for nitrogen.

Image:  ASCENT propellant.  R. Fair / Media Fusion.

--

Refs.

Berezhnoy, A.A., Kozlova, E.A., Sinitsyn, M.P., Shangaraev, A.A. and Shevchenko, V.V., 2012. Origin and stability of lunar polar volatiles. Advances in space research, 50(12), pp.1638-1646.

Cannon, K.M., 2021. Accessible Carbon on the Moon. arXiv preprint arXiv:2104.13521.

[LMT]

Breakeven analysis for lunar LOX and other marginal Mars mission lunar-depot scenarios will require accuracy typically missing from BOE calcs.

A nice tool with useful accuracy is MIT's SpaceNet.  See the various lunar and Mars mission applications as starting points for constructing a lunar Mars LOX breakeven analysis, or a more favorable LOX+LCH4 analysis.

[davis_404]

Regardless of how, landing Ships with tanks depleted of LOX and still carrying a large remnant of methane makes sense. Both Mars and especially Luna (yeah, I'm reinstituting that name) have oxygen on hand. LOX is far heavier than liquid methane. Rather than re-fueling, think of re-LOXing.

[Twark_Main]

A lot of these considerations were already discussed in the Purgatory'ed discussion. I had hoped people would "hold off" discussion until we got those posts moved over (hence "placeholder"), but I guess I should have made that more explicit. I would have locked the thread, but I would've needed a moderator for that too.


Moderators, can we get those posts moved over (ie the intent of this thread) so we don't have to re-hash a lot of these calculations and discussions again? I used the Report function and also set a PM to zubenelgenubi, but no luck so far.

Anyone know a better way to get some much-needed mod attention?

[LMT]

we don't have to re-hash a lot of these calculations

Since we're setting the stage (again), I think a breakeven calc on some lunar depot for Mars flight should use a reference at least as robust as that of MIT's SpaceNet.  Margins are often thin in such scenarios, and simplifying assumptions can flip breakeven artificially.

If a poster doesn't care for SpaceNet, some reasoning should be given:  e.g., show how a spreadsheet gives an important metric that matches a SpaceNet example, in a certain Mars launch window.

[Twark_Main]

we don't have to re-hash a lot of these calculations

Since we're setting the stage (again), I think a breakeven calc on some lunar depot for Mars flight should use a reference at least as robust as that of MIT's SpaceNet.  Margins are often thin in such scenarios, and simplifying assumptions can flip breakeven artificially.

If a poster doesn't care for SpaceNet, some reasoning should be given:  e.g., show how a spreadsheet gives an important metric that matches a SpaceNet example, in a certain Mars launch window.

SpaceNet (which appears to use a simple delta-v map simplification) is actually far less robust than the calculations TheRadicalModerate performed (which use an actual Lambert solver).

I just wish you could see 'em!

Anyway, I don't think it's necessary (or desirable) to pre-establish some sort of complex set of "standards" here. If someone makes a math mistake, just point it out to them.

[LMT]

we don't have to re-hash a lot of these calculations

Since we're setting the stage (again), I think a breakeven calc on some lunar depot for Mars flight should use a reference at least as robust as that of MIT's SpaceNet.  Margins are often thin in such scenarios, and simplifying assumptions can flip breakeven artificially.

If a poster doesn't care for SpaceNet, some reasoning should be given:  e.g., show how a spreadsheet gives an important metric that matches a SpaceNet example, in a certain Mars launch window.

SpaceNet (which appears to use a simple delta-v map simplification) is actually far less robust than the calculations TheRadicalModerate performed (which use an actual Lambert solver).

   Posters might start analysis with various SpaceNet Mars mission examples, which apply the time-expanded network model (with time-varying transfer feasibility), multi-year mission delta-v requirements, and other physical constraints.

[TheRadicalModerate]

Since it appears that the original sub-thread has been black-holed (which is completely unjustified and extremely freakin' annoying; we may have been slightly off-topic but it was a good-faith and interesting discussion that should not have been capriciously bit-bucketed--very poor moderation decision), let me reattach my cartoons for some of the relevant cases:

1) LUNOX used solely for Earth-Moon shuttling.  Note that this assumes that all LCH4 is added in LEO, and that the tanks aren't re-sized.  Max payload for this condition is 135t.

2) No LUNOX used for departure to Mars with v∞=6.91m/s. I'm going to use this as a baseline, simply because it's what I computed as what seemed to be the quickest transit time possible in the 2028 window, but the arrival v∞ may still be a little hot.  This is just a vanilla direct departure from LEO.  Payload assumed to be 150t.

3) Departure from the surface of the Moon, with the assumption that the Starship launches into a C3=-2 TEI, then does the final departure burn for Mars from perigee.  Payload assumed to be 150t

4) Departure after refueling in NRHO, dropping into the same C3=-2 orbit and then doing the burn at perigee.

5) One that I have not yet done would be departure after refueling directly in C3=-2.  Twark and I disagree about the risk of using extreme HEEOs for Mars departure.  I think that any snafu with refueling requires a lot of delta-v to move the argument of perigee for the next orbit's attempt, and he doesn't.  I'll try to get this one once I've recovered the mindset.

Bottom line on these:  If you're propulsively launching LUNOX off the Moon, either in the target vehicle or in a special tanker that fills a depot, LUNOX has to be awfully cheap before this makes sense.  On the other hand, if you have a mass driver that can deliver LUNOX direct to a depot, then the economics may work out.

I don't expect this to be in routine use until a lunar metals industry is pretty mature.  Since metals reduction produced copious amounts of oxygen, this seems to be the point where LUNOX might make sense.

Again, I haven't really looked at these since they were taken down the first time.  I was reasonably sure that they were accurate back in the day.

[LMT]

LUNOX has to be awfully cheap before this makes sense.  On the other hand, if you have a mass driver that can deliver LUNOX direct to a depot, then the economics may work out.

So long as lunar EM launch is a dedicated multibillion dollar infrastructure, its LUNOX payload can't be "awfully cheap", relative to LOX from shared Earth infrastructure.  Roesler 2013. 

Note the EM launcher's daily delivery:  240 kg.

--

TheRadicalModerate:  Can you compare your case (4.) numbers against SpaceNet prediction for equivalent 2029 LUNOX TMI?

--

Refs.

Roesler, G., 2013. Mass Estimate for a Lunar Resource Launcher Based on Existing Terrestrial Electromagnetic Launchers. Machines, 1(1), pp.50-62.

[TheRadicalModerate]

Note the EM launcher's daily delivery:  240 kg.

You'd have to do better than that.  If you want to surge 10 flights to Mars through a 2 month window, you'd need about 65t per day.  However, I don't think you're constrained by existing electromagnetic launchers.

Remember that my basic position on LUNOX is that it's silly for a long time.  The only way it ever makes sense is if its cost to orbit is a fraction of the launch cost from Earth to LEO.  I don't see how you get that without some kind of high-volume mass driver.  If that isn't possible, then LUNOX won't happen.

[libra]

Use Al-lox SSTOs manufactured on the Moon. 285 seconds specific impulse is not good... by Earth gravity standards. On the Moon with 1/6 the gravity, it's plenty enough to do wonders. Lunar orbit from the surface is 2400 m/s; Moon escape velocity from the surface is 2700 m/s - vs 9300 m/s and 11200 m/s for Earth !

How hard would it be to manufacture on the Moon
a) aluminum tanks
and
b) something akin to the Lunar Module ascent engine (except burning aluminum with LOX) ?

What's more, once in lunar orbit the lunar SSTO can be used as a LOX depot. Rinse, repeat.

9.81*285*ln((50+30)/(2.5+30)) = 2518 m/s - so 30 mt of LOX for a 50 mt SSTLO: Single Stage To Lunar Orbit.

EDIT: at least far better than 240 kg, and I'm closer from the aforementionned 65 mt !

[TheRadicalModerate]

Use Al-lox SSTOs manufactured on the Moon. 285 seconds specific impulse is not good... by Earth gravity standards. On the Moon with 1/6 the gravity, it's plenty enough to do wonders. Lunar orbit from the surface is 2400 m/s; Moon escape velocity from the surface is 2700 m/s - vs 9300 m/s and 11200 m/s for Earth !

How hard would it be to manufacture on the Moon
a) aluminum tanks
and
b) something akin to the Lunar Module ascent engine (except burning aluminum with LOX) ?

What's more, once in lunar orbit the lunar SSTO can be used as a LOX depot. Rinse, repeat.

9.81*285*ln((50+30)/(2.5+30)) = 2518 m/s - so 30 mt of LOX for a 50 mt SSTLO: Single Stage To Lunar Orbit.

There are three big problems with your analysis:

1) Even with the heavy delta-v penalty to get to LEO, that only costs a lot of prop, which is essentially free on Earth, while it costs thousands of dollars per tonne to make on the Moon.

2) A tanker going from Earth to a parking orbit suitable for staging to Mars (I've been assuming something TLI-like, 385,000 x 400km) only has to spend about 250m/s to land, because it can aerobrake.  A tanker from the lunar surface has to spend about 3000m/s to land.  Even with the ship being empty, that still eats into whatever advantage you had from staging from the shallower gravity well.  (For hydrolox, the Moon-staged total prop is still a bit less, but for methalox, it's 50% more than from LEO.)

3) Moon-based launchers are going to have shorter lifetimes (due to limited maintenance opportunities and dust corrosion), and much higher operational costs (due to most maintenance having to occur in vacuum, with techs that have to live on-site, which costs hundreds to thousands of times more per tech-hour than it would on Earth).  I'd be surprised if a lunar launch cost less than 20x what a terrestrial launch cost.

You can whittle down the cost premium of operating from the Moon, but only if you have a massive industrial base, with thousands of humans, living at least 80% self-sufficiently.  That's well over what I consider the dividing line between reasonable extrapolation and science fiction.