From (email me there please)… This is a sub-site to main site at

This web page last updated 06 August 2019


Designs for Recycling Rocket Exhausts on the Moon or Mars


Abstract / Pre-Summary

            This sub-page to is meant to describe methods for recycling rocket exhausts.  “Volatiles” (such as water, carbon dioxide, carbon monoxide, and other gasses as might be found in rocket exhaust) are precious on Mars, and even more so, on the Moon.  Methods described here could be used to recover some significant percentage of such “volatiles” (gasses) in the exhaust plumes of landing (or launching) rockets, on any airless or near-airless “heavenly body”.  After collecting such cooled-back-down (liquid or solid) volatiles, they can be (with the input of energy) recycled for uses such as rocket fuel, drinking and sanitation, and industrial and agricultural uses.

            Two primary versions are described here:  Revising an existing cave (such as a “lava tube”), or building an exhaust-recycler “from scratch” (digging or building your own caves or tunnels).  In either case, the rocket exhausts are caught in cold voids shaded from the heat of the sun.  There, they are allowed to cool, condense, and be collected for recycling.

             As with other sub-pages of , the intent here is to “defensively publish” propulsion-related (and “misc.”) ideas, to make them available to everyone “for free”, and to prevent “patent trolling” of (mostly) simple, basic ideas.



            The virtues of recycling rocket exhausts hardly needs to be expounded upon (at a remote moon or planet which lacks many volatiles, that is).  The Earth’s Moon is a first and primary target of the discussions here, but Mars would work as well.  The Moon lacks significant amounts of carbon (unlike Mars, where carbon can be harvested from the thin carbon dioxide atmosphere).  So on the Moon, capturing carbon-dioxide-containing exhausts would be especially advantageous.  This would apply, for example, to rockets that use methane fuel, such as SpaceX’s “Starship”.

            Recycling volatiles in not the only advantage of the designs presented here…  The other main advantage is eliminating (or at least suppressing) the problem of rocket exhausts blowing dust all over everything and everyone, close to the landing / launching pad.  For details about that, please see, for example, , “Apollo taught us that landing on the Moon is a dusty nightmare.”


Re-Purposing a Cave (Lava Tube)


            Links concerning caves (empty lava tubes) on the Moon and Mars can easily be found.  See, for examples, , , and .  Sure, we could live in caves, protected from micro-meteors, radiation, and temperature extremes.  But why not also use caves for recycling rocket exhausts?

            OK, so, then, time for some drawings!  We COULD take a natural opening (sky-light style) to a cave, and build a bridge (suspension-style or otherwise) over the hole, and put a landing (or launching) pad in the middle of the bridge.  A metal grid (with plenty of holes or voids) could be used for the landing-pad.  If one fears for the metal grid melting or otherwise degrading over time, under the mechanical and heat-assaults of the rocket exhaust from many-many rocket landings, then the top-most surface of the metal grid could be clad with high-temperature-tolerant ceramics.  Ceramics should be able to be easily sourced locally on the Moon or Mars, and replaced as they wear out or break.

            Simple, yes, but to be sure, here is a drawing of the “grid” surface of the landing grid (surface of landing [or launching] bridge)…  “Open grid steel decking” is a Google search-string that will tell you a LOT about associated matters!  Also see for example.  The metal that we’d use for our high-temperature-resistant application might be titanium (somewhat expensive), or stainless steel (310S alloy to be specific), for example.  If good-enough high-temperature-tolerant metals are used, a ceramic top-cladding material would probably not be needed.

            On, to the basic drawings!



Figure #1


            Figure #1 (above) needs no more comments, so let’s move right along to #2…



Figure #2


            Moving right along to Figure #3, let us draw a side view of a cave (lava tube).  The idea of using a bridge (covering part of a natural sun-roof-type opening at the top of the cave, with a rocket landing-launching pad, with a flame-permeable grid) scarcely needs a separate drawing…  So the below drawing shows an artificial hole has been cut into the cave, for the landing-launching pad.  A natural opening (if one exists) can be used for other uses, such as human access.  Or, a single natural or human-cut opening can be double-used.



Figure #3


            So, fairly obviously, the exhaust gasses will be retained in the sun-shaded cooling temperatures of the cave, where they will condense into liquids and even solids, falling out as they cool.  They can then be collected and recycled.

            The following comments will apply, whether the cave is natural or artificial:  Recycling the condensed exhausts will be made easier by making the cave surfaces (or at the very least, the cave bottom) lined, such as to be impermeable by the exhausts (whatever their phase-state may be).  One way to affordably fashion such impermeable surfaces (on the Moon especially) may be to gather fine Moon dust (regolith), cover the surface to be lined, and then microwave it.  See , for example.  3-D printing methods might also be used, for the manufacture of blocks for pavements or other uses.  See for example.

            Mooncrete” (AKA lunarcrete) could also be used (for sealing the landing surfaces, and cave walls, ceilings, and floors).  See .  Perhaps such versions of concrete might be sprayed on cave rock-walls and ceilings in a manner similar to the spraying of pool-wall “gunite”.  See

            Artificial barriers inside the cave can be used to separate areas, one from the other, for exhaust recycling v/s human uses.  The area (volume) for exhaust recycling will need to be optimally sized.  The attributes of such an area (volume) could be validly compared to what has been learned on Earth, with launch-pad “flame trenches”.  See for example.  If the volume (for exhausts) is too low, the gasses will “blow back” out of the launching / landing pad, going to waste, and possibly even endangering the spacecraft.

            The Moon’s (or Mars’s) outermost surface surrounding the landing-launching pad would ideally be smooth-paved (via microwaved regolith, or other methods) to reduce having dust being blown all over the local area (around the pad, by rocket exhausts).  But such ideas have already been described at .  Optimally, such pavement would not only prevent regolith (lunar dust) from being blown all around, it would also keep our being-captured exhaust gasses from seeping through the rock, and escaping.

            Above and beyond that, though, here’s another idea:  The volatiles (gasses) that we’re trying to capture here, might escape from our recycling efforts, at the uppermost stretches of the opening above our cave, whether artificial or natural.  The volatiles might escape through cracks in the rocks, especially closest to the surface (where the exhaust gasses are hottest, and where residual heat energy from the sun lingers in the rocks and dirt, even at night).  So therefor, for at least a few yards or tens of yards downwards, we might want to surround the landing pad with drilled holes, and seal the cracks in the rocks with a “grout curtain” (injected concrete).  For details about what a “grout curtain” is, use “grout curtain” as a search string, “Google” or otherwise.  (I’m not finding very highly relevant descriptive links here, for “grout curtain”, so I’m giving up for now).

            In other words, and in more detail, the uppermost rocks all around the landing pad (in a doughnut-shape or torus or annular ring) will almost certainly be like a porous sponge, with cracks and voids that will absorb the gasses that we’re trying to capture and recycle.  To prevent their escape, we’ll pave (or “paint” with mooncrete gunite, or some such) the surrounding surfaces, to seal them as best as we can.  The uppermost surfaces of the vertical hole (right underneath the landing grille or flame grid) may be subjected to such frequently-repeated highly destructive temperatures, that sealants such as mooncrete gunite might not hold up very well.  That’s why we back off a few yards radially outward, and drill down some holes into the “doughnut” here, and squirt in a “grout curtain”.  Now our system might still not work quite perfectly…  Nothing ever does…  But we’ve made several hopefully-practical and affordable improvements.  Parts of the “doughnut” might soak up some gasses, but once saturated, they should soak up no more…  And allow very few gasses to “sneak around the corners” to escape!  Further below, see a repeated drawing (from above) with changed labels, to clarify these matters.

            Preferentially landing and launching in the middle of the night (with regards to the Moon or Mars or otherwise, but especially with regards to the Moon, with a 2-week, very long and cold night) will keep temperatures low, and will help you to recycle exhaust gasses.  That’s pretty obvious, but let it at least be mentioned here…  Done!



Figure #4


            For any more clarification (or drawings) on the above (or anything else here), please email me at .

The above are the at-least-vaguely-plausible suggestions.  Before we move on, to describing totally custom-cut (artificial) “caves” for recycling exhaust gasses, let us briefly veer off, into the less-plausible (IMHO, In My Humble Opinion) options.  But, as a reminder, we’re here to fend off the “patent trolls” by “defensively publishing” ideas, here, so that they can NOT be patented!  ALL ideas (including implausible ones) are potentially to be “freed up” from the patent trolls, here!

            The exhausts-cooling-down-cave’s cold surfaces (especially at the points where the hot gasses are optimally hot, not hot enough to destroy the specially lined surfaces, but hot enough to recover some useful energy) might be lined with thermal-differential power-recovering materials (or “metamaterials”) such as are described here: .  “Thermal metamaterials” and “heat energy harvesting” might make good internet search strings.  Such “tech” MIGHT make it economical to recover energy from the thermal difference between the hot exhaust gasses, and the cold cave walls.  See also and (The latter being “A FlowerShaped Thermal Energy Harvester Made by Metamaterials

“).  Also .

            Also, it MIGHT (perhaps just barely?) make sense to “heat pump” the heat out of the cave, into the cold night, on the Moon, or on Mars, to cool or “pre-prep” the cave, to do a better job of cooling down the exhaust gasses.  The thin (Mars) or non-existent (Moon) “atmosphere” wouldn’t help us much via “heat convection”, but “heat radiation” might do the trick.  I have no exotic or brilliant ideas to offer here.  See “Peltier effect” for grins.  See for that...  Alternately, we could heat-pump the heat from the cooled volume (gasses capture volume) of the cave, to human-habitation areas (volumes), when heat is needed in the habitation volumes.

            OK, on, then, to describe what things might look like, if we custom-carve a “cave” (or set of caves) for this purpose, of collecting and recycling exhaust gasses…


Custom-Cutting Caves for Exhaust Gasses Recycling


            Let’s first describe a “deluxe” custom-cut hole or cave.  We’re assuming here that we use tunnel-boring machines, automated (roboticized) as much as possible.  Bigger holes, and more holes, of course, will cost more.  A “deluxe” hole might be very wide and very deep, with perhaps one smaller, slanted access hole, for getting to the bottom and retrieving (or pumping out) the gathered (condensed) volatiles.  Also shown here is an optional (probably powered, not passive) one-way shut-off valve.  This would be located far enough down to let the exhaust gasses cool off a bit, first, before hitting this large one-way gas-flow valve (super-hot gasses will endanger our valve).  Open the valve for launches and landings, of course, but then shut it down after the gasses are captured (prevent excess back-flow and waste of the gasses).



Figure #5


            So then, dropping back from the “deluxe” system (assumed to be very expensive, and impractical for a long time, for all but the busiest spaceports), let us describe some less-expensive implementations and options.  I assume here that smaller-diameter tunnels and holes are more affordable.  I do know that more-gentle fluid-flow turns (whether we are turning the flow of liquids or gasses, matters not) will impede fluid-flows less drastically.  What I don’t know, is what kinds of tunnels can be bored most cost-effectively.  So several options are shown here.

            First off, let’s show what the top layer of a more-affordable “gas recycling” spaceport might look like.  Assume that we start with a fairly flat surface (of the Moon or Mars or other).  We pave or otherwise seal a flat landing-launching surface, without bothering with ANY kind of (or with very minimal) provisions for gently re-directing exhaust gasses from the mode of being propelled downwards, to the mode of being propelled at a right angle from downwards (that is, to then being directed towards outwards from the pad-center).  After being re-directed from travelling downwards, the exhaust gasses will have turned a right angle, and travel into artificial “caves” directed outwards, radially all around the pad.  Each cave-entrance will be constructed out of blocks fashioned out of “in situ” gathered materials, mostly.

One example of such cave-entrance construction might look like the classical “arch with keystone” design, with insulating top-fill between (and on top of)  the tops of the arches.  Note that grout curtains and sealants (mooncrete gunite or other) could apply to some of the drawings below, but have already been discussed and diagrammed, and so, are omitted (to cut clutter) from here on down.

So here’s what we might see, if we were sitting in the middle of the pad, looking towards the cave entrances…



Figure #6


A top-side view or bird’s-eye view (OK, no birds are expected on the Moon or Mars, so a rocket-drone’s view, then) of the radially arranged gas-capture caves might look like this:



Figure #7


Your humble author here hasn’t a clue about the following:  How expensive will it be to carve artificial caves (tunnel-bore) under the Moon or Mars?  How will that compare to the costs of building caves almost purely above the surface?  How far do we have to go down (or how thick do we have to cover our artificial surface caves?) to get the cold temperatures that we need?  How expensive is it, to tunnel-bore sharply-turned tunnels, either underground, or at the surface?

Depending on the answers to the above, any number of configurations could be used.  The tunnels (artificial caves) shown above, could run parallel to the surface, for whatever total length is needed to cool the gasses.  This is so simple, as to not be diagrammed.  Or, the tunnel could go out a short ways, and then take a SHARP turn (or a SHALLOW turn) as it dives down deep under the rock and dirt, then comes back out.  Like this…  Note, only one of many outgoing “octopus arm tunnels” is shown, going out from the central landing pad, for simplicity, but the ideas should be clear…



Figure #8


Simply for illustrating some fairly obvious ideas, in the name of completeness, let’s provide a few more discussions and drawings.  The below is a view from above, of one of many radial surface tunnels built around the landing-launching pad.  The first (innermost) tunnel-or-cave chamber should be long, to make sure that there’s plenty of volume to capture the gasses (and to prevent wasteful and dangerous blow-back).  The entire tunnel could be straight, but is shown here as sinusoidal, partly just to fit it on the page.

While we’re building a tunnel (or cave) anyway, we might as well be efficient, and put it to multiple uses.  After the cooling-down (large volume) compartment, we place a wall, and then a greenhouse.  If practical (or when) cooled-down gasses are ever sufficiently dense for this to make common sense energy-expenditure-wise, the gasses can be pumped (presumably from lower pressure to higher pressure) from the cooling chamber to the greenhouse chamber.  If the gasses can be collected from the floor of the cooling chamber in LIQUID form (after condensing), they could be pumped to the greenhouse more conveniently.  Periodic excursions (by robots and-or humans) to collect solid-phase condensates, and move them through an airlock to the greenhouse chamber, may make sense, also.



Figure #9


The following may be obvious, but let’s add it for completeness:  8 or 12 or “N” tunnels or caves entrances arranged radially around the space-port pad makes sense, for not immediately impeding hot gasses from flowing freely.  The “plumbing” details are highly variable, though.  8 or 12 separate greenhouses and human-use areas may make little sense.  Tie the chambers together and combine them to smaller total numbers may make a lot more sense (but still preserving the radial arrangement of at least the first parts of the cooling chambers, for not impeding the flow of the high-speed, high-pressure exhaust gasses).  The above is illustrative only.

            Running the gasses through a greenhouse, first, before human use, allows plants to pull out excess carbon dioxide (and convert it to oxygen).  Greenhouses can be maintained by a mix of human and robot labor.  Humans (in the greenhouse) may need to wear oxygen breathing gear if the carbon dioxide (or other gas levels, or smell) is dangerous or unpleasant to humans.  Note that not only plants, but also soil bacterial, fungi, and assorted microbes can clean toxins and bad smells out of air.  This is especially true of the air is forced (pumped) through soil beds deliberately.  Biosphere II showed this to be true.  I have not looked for the very-very best link to show this, but here is one for starters: .

            After lingering in the greenhouse for a while, excess air and-or water can be removed, cleaned and conditioned some more, and used for humans, of course.  Food grown in the greenhouse will need to be monitored for contaminants.  Only fairly clean-burning rocket fuels should be allowed to feed this kind of scheme!

            Some of the drawings above show a gated one-way valve to prevent backflow, fairly close to the hot exhausts.  This is probably the most plausible and practical idea.  However, other drawings show the tunnels or caves as completely open.  This is also possible (as is also, a mix of the two types of tunnels, even at the same spaceport).  If that latter idea is used, then (probably best) very close to the bottom-most part of the cave, it might be wise to place two types of “rubble piles” (or permeable filters)…  One made of potassium hydroxide or lithium hydroxide, or other suitable chemical for capturing carbon dioxide, and another pile (of salts or other desiccants) for absorbing water.  The rubble piles would impede the out-flow (and waste) of the gasses, at the other end of the tunnel or cave.  These rubble piles (or filters) would also be periodically re-processed to extract the volatiles.

            Now let’s make an at-first-apparently-irrelevant detour, but then circle back around and show its relevance.  There’s a very-very thin atmosphere on the Moon, if we can even call it that.  However, as human activity ramps up on and around the Moon, it will get thicker.  Here are the sources:

            ‘1)  The solar wind contains atomic nuclei, especially of hydrogen and helium, but other elements as well.  Some of these bounce around the moon a while, and can get caught in “cold traps”, whether man-made or natural.  (They will also capture electrons from the solar wind, or from the surface of the moon, to become mostly-neutral atoms).

            ‘2)  Impacting asteroids and comets deposit volatiles as well.

            ‘3)  Future rockets, satellites, and “space tugs” will emit gasses as well.  Including conventional chemical propulsion, of course!  Nuclear-thermal tugs may emit hydrogen, and cold-gas maneuvering rockets may emit nitrogen (for example) if they originated on Earth.  Moon-made cold-gas rockets and satellites (made using “in situ” locally sourced materials) might opt to use oxygen instead…  Nitrogen is near-non-existent on the moon, but oxygen can be separated from very common silicon, aluminum, iron, titanium, and etc. (other metal) oxides.

            ‘4)  Heavy wheeled moon vehicles (see for example) will transport humans, moon rocks, and supplies to and from the Moon’s “outback” regions (as the Moon is explored for scientific, mining-prospecting, and other reasons), and in between bases.  They will scrub their carbon dioxide out of their air, using potassium hydroxide or lithium hydroxide, or other scrubbers.  They may not want to bother to bring all of their spent carbon dioxide back home…  They may want to carry more rocks and supplies instead.  If practical, they will re-process their scrubbers instead, and dump their carbon dioxide overboard.

            ‘5)  Another method of exploring the Moon might use rocket-based “hoppers”.  When departing from our spaceport, we can recycle rocket exhaust as usual…  So burn a near-perfect match of, say, methane and oxygen, when departing…  We’ll capture most of it back anyway!  Get your highest “specific impulse” as is practical, out of the fuel, when departing.  When “hopping” around in the outback, though, we might settle for lower specific impulse (burn cooler), and burn way oxygen-rich, because oxygen is far more available on the Moon (for reaction mass).  So we’ll be adding more oxygen to the Moon’s way-thin “atmosphere”, if we do that.

            Also plausible idea (when in rocket-hopping mode) is this:  Start your burn with a normal mix of methane and oxygen.  (When starting a retro-burn to come out of free-fall, this will help “settle the masses” in your tanks, including a tank of finely powdered aluminum…  Read on…).  Now throttle back on the methane, but keep the flame lit so that the next (other, supplemented) fire will stay lit also.  Dump un-oxidized metallic finely-ground aluminum powder (extracted locally) and excessive amounts of oxygen into the fires (rocket’s combustion chamber).  Burn oxygen-rich again, here, is probably quite wise.

            For these reasons (and probably more, especially as human activities ramp up), the Moon’s very thin atmosphere may become a wee tad richer.  Human-added oxygen atoms will snag hydrogen atoms from the solar wind, adding weight to the atoms (making water), which will be retained in “cold traps” more efficiently.  So the below-listed ideas might JUST BARELY start to make sense, as time goes by.

            If you’re going to bore a tunnel down and back up, we might as well use the other (non-rocket-pad) end of the tunnel to capture some of the Moon’s thin atmosphere, especially at night.  It might even make sense to spend a tiny bit of extra money at this far end, and carve a mini-funnel or mini-crater to improve the “cold trap”.  Especially in the middle of the lunar night, if and when the bottom-of-the-cave cold trap is truly cold (there’s not hot or warm gasses in there cooling down, right now), we open up the far-end gate.  See the drawing below.



Figure #10


The far-end gas-flow gate needs to (when shut) prevent gas-flow in both directions, obviously.  Over a long-long time, the funnel-bottom may collect volatiles outside of the far-side gate.  These can be collected with some extra trouble, every once in a great while.  Or, we could put TWO gates at the far end, so as to capture volatiles at the far end, even while cooling down a batch of hot gas at the bottom.

The entire spaceport and tunnels and-or caves should ideally be located inside a moderate to large-sized crater.  This (the crater walls) will protect the rest of the Moon (or at least, other nearby bases or settlements) from any escaping rocket-blast-blown Moon dust (regolith).

Excavating the far-side funnel is probably more trouble than it is worth, but it is documented here in the name of thoroughness.  If these funnels are dug, the resulting extra “fill” debris could be used to patch any low spots in the surrounding crater walls.

            Falling into the same category of probably-not-worth it, we could erect, around the rim of the enveloping crater, “snow fences”.  As snow fences on Earth slow down the wind, causing snow flakes to drop out, our “snow fence” (made of slats or poles, and voids) could slow down the extremely diffuse “atmospheric winds”, and help us catch stray volatile molecules.  Structural integrity to countervail against gravity isn’t as important on the Moon as on Earth, and the “winds” will be weak to the vanishing point, so we can heavily tilt our “snow fence” around the crater rim.  Like this:



Figure #11


Here are concluding remarks for this sub-page, some of which should be obvious:  Recycling rocket exhausts makes sense, only on fairly highly-trafficked, well-developed spaceports.  These kinds of ideas won’t likely be practical till (?) the late 2030s at the soonest, or some time after that.  The ideas here will be more useful, the further away a Moon spaceport is located away from the Moon’s poles, where there strongly appear to be deposits of volatiles.  And SOME of the ideas listed here, could apply to processing lunar-pole deposits, in the same manner as (or combined with) the processing of recycled rocket exhausts.

If the ambient-gasses-collecting funnels (see above) and “snow fences” are used, then one more step or set of steps MIGHT make sense as well:  Some sort of electric charges, magnetic fields, or combination of both might be set up, to attract more charged particles from the solar wind (to neutralize and capture them for recycling, of course).  That might be a bad idea, though, for increasing electromagnetic noise for electronics, and radiation hazards for humans.

Also, I really have no idea what such an arrangement might look like.  A Moon-orbiting electromagnet that is ONLY turned on at the right time and place, perhaps?  That’s one guess of mine!  If you have information or ideas along these lines, please send them to me at

Perhaps relevant to the above, is an idea that I have seen, which is to use the Mars “Lagrange Point” of “L1”, with respect to Mars’ orbit around the sun.  Set up a large electromagnet there.  The “magnetotail” of the magnetic field will now envelop Mars, protecting Mars’ atmosphere from continuing to lose yet more gasses to the solar wind (as it has been doing ever since Mars’ magnetic field collapsed 4.2 billion years ago, or so).  See   NASA proposes a magnetic shield to protect Mars' atmosphere”.

            However, this scheme DEFLECTS rather than COLLECTS the solar wind.  So in our scheme, I think we might want to put TWO electromagnets in orbit around the Moon.  Turn them both on, only when our collection-base is caught in between the two electromagnetic satellites (and also, only when there are no humans on the Moon’s surface, exposed to extra radiation).  Now, between the TWO electromagnets, more solar wind will be “shepherded” down for collection.  That’s my best stab at it…

I have no special expertise or any more plausible ideas concerning any associated matters here, so I will sign off at this time.  This concludes my ideas as of this time.  Once again, comments or questions (or idea contributions) are welcomed at


Stay tuned…  Talk to me!


Back to main site at