Smog Solar Engine

guest blogger john hare

 I fail to understand the attraction of the complex nuclear and electric engines for deep space maneuvering when solar thermal promises to be so much simpler and higher performance. A few simple modifications to the normal methods of heating the working fluid and nuclear thermal becomes an under performing, over politicized dead end.

There has been some work in laser launch on letting the beam hit particles in the exhaust rather than the solid portions of the engine itself. By constantly reheating the expanding gasses with the heat pumped into the smog particles, the natural tendency of expanding gasses to cool is counteracted so that a relatively cool exhaust through the throat has an unlimited expansion ratio. Unlimited by physics, not by optimum mass and performance. The smog/water propellant for laser launch could exceed the Isp and T/W of the SSMEs by a good margin. Smog /hydrogen Isp can be toward 2,000.

Laser and other beamed propulsion has to contend with the financial aspects of beating established techniques for Earth launch. The business case doesn’t close unless extremely high launch volumes are projected due to the high capital costs. Solar in space can use the same techniques without the same problems faced by lasers. Propulsion mass and propellant must be lifted anyway, the most efficient means of using them should score high in a trade study, and hardware costs should be well below most of the competitors.

smog solar

The sunlight cannot be focused to a tight spot right into the ‘combustion’ bell as a laser could. Even if the mirrors were dead astern, they would be in the exhaust path which would be most unfortunate. By using one side of the linear aerospike concept, sunlight can be focused on the expansion plume without the mirrors being close to the exhaust path at all. Most of the light will hit the smog particles in the expanding plume for some really intense afterburning effects. The exhaust can be in the thousands of degrees even in the exit plane of the nozzle.

The nozzle will get some heating from the sunlight that ‘leaks’ through the smog as well as from the exhaust gasses. Enough heat will be absorbed to run the expander cycle propellant pump, as well as enough heat for the initial expansion through the throat to supersonic velocities. The solid portions of the engine will never be subjected to the temperatures of nuclear thermal or most competing solar thermal concepts. With propellant use in the grams per second, a large number of very small throats will be requires to feed the large expansion ratio. This is the one side linear aerospike layout with perhaps dozens of throats.


Most desired thrust directions in deep space missions will be at close to 90 degrees from the sun. Either accelerating or decelerating will be modification of a solar orbit. This allows the mirror to be in a constant relative position to the ship off to one side and slightly behind the engine section. While very large, the mirror is not necessarily massive. It is however, awkward and off center. With the low thrusts that will be used, less than 1 m/s, the mirror will never see more than 1/10 gee. The supporting truss can be a gossamer affair with almost fishing line size cables for mirror support. The counterweight can be almost anything useful to the mission.

If coasting periods are sufficiently long, the truss might be adaptable for a centrifugal arm to provide some gravity to a small module.

The following two tabs change content below.


I do construction for a living and aerospace as an occasional hobby. I am an inventor and a bit of an entrepreneur. I've been self employed since the 1980s and working in concrete since the 1970s. When I grow up, I want to work with rockets and spacecraft. I did a stupid rocket trick a few decades back and decided not to try another hot fire without adult supervision. Haven't located much of that as we are all big kids when working with our passions.

Latest posts by johnhare (see all)

This entry was posted in Uncategorized. Bookmark the permalink.

21 Responses to Smog Solar Engine

  1. MG says:

    It is non-obvious to me that you will get the heating you anticipate getting.

    First, what is the actual composition of the “smog”? Carbon particles and water droplets?

    *If* you get adequate radiative absorption by the smog particles, then will they re-radiate at frequencies that couple with the other exhaust products? Will the exhaust have sufficient density to equilibrate the translational, vibrational, and spin temperatures? If so, at what expansion range does that stop being the case?

  2. A_M_Swallow says:

    I like solar thermal.

    I can see a few problems with your design. The smog particles will block the small throats. The size of particles and size of throat will have to be adjusted to minimise this. The chamber will have to be cleaned after every flight.

    A maximum acceleration of 0.1 g. That is a tantalizing number. Lunar gravity is 0.1654 g (1.622 m/s2). Get the acceleration up to 0.2 g and you have the engine for a reusable lunar lander.

    On the moon ISRU LOX is available to act as a propellant gas. The smog could be made from raw regolith, aluminium, iron, silicon or a compound. Powdered aluminium can be burnt for extra thrust during landings and take-off.

  3. johnhare john hare says:


    I’m handwaving the particles as being similar to the solutions the laser guys found. That in turn is based on a very short hallway conversation with Jordan Kare at a Space Access conference several years back. I don’t know what the optimum particle composition is, just that they have to be very very small. My assumption is that if your pour enough energy into the exhaust volume, it will get and stay hot enough. It would seem that mirrors can be fairly light, and can therefore be increased in size with fairly little pain.

    I also have a vague recollection of Henry Spencer saying that hydrogen was opaque to some wavelength at sufficiently high temperatures. Since the competing technology is nuclear thermal at ~900? seconds, getting enough power into the gas or the particles seems simple enough. Total temperature can be well below chemical rocket normal and still give impressive Isp with the super light H2 exhaust with impurities.

    It would be interesting to figure out just where the expansion becomes too great for further heating.


    The particles will probably be nanoparticles (or smaller?) mixed into the H2 in relatively miniscule mass quantities. I don’t know the best composition or even proportion of the particles. MG hints that water vapor might work. If the impurities can be a liquid or even a gas, then throat blockage and cleaning becomes a non-issue.

    I don’t think this could ever be a lander engine. That mirror is going to have to be fragile to keep the mass down for it’s primary purpose. Perhaps leave the mirror in orbit with a lunar laser to supply landing and launching energy to a core engine.

    ISRU LLOX is a good point. Even with about a third of the Isp of H2, availability and shipping costs make it an attractive alternative. Engine design would have to accomodate an oxydizing propellant though, may take a different engine entirely. If various ISRU options can be used though, it opens up possibilities. LLOX, Martian CO2, or asteroid volitiles as available. Since even a low Isp is attractive if propellants are available everywhere in forms that lead to high mass ratios.

  4. A_M_Swallow says:

    Grinding down local materials to form dust is not difficult, we can even use ISRU “stones”. The motors and power would have to come from Earth.

    A field of mirrors on the moon is a possible alternative to a lunar laser, similar to the one heating the solar tower in Spain.

    Ceramics can take high temperatures without oxidising, some are very hard.

  5. A_M_Swallow says:

    If the nozzle is only on one side will this result in the vehicle having large rotational forces?

    Can a very wide nozzle be used with two mirrors? One on each side to give symmetry.

  6. johnhare john hare says:

    The symmetric mirrors would be a much better layout from both a thrust alignment and structural standpoint. A symmetric true linear aerospike nozzle layout could then be used.

    I’m fairly certain the Lunar field of mirrors won’t work at the long ranges from the vehicle required. The rapid refocusing with distance change and aiming would be a problem also.

  7. Tim says:

    Hi John. Could an ablative throat lining act as a source of particles?
    If you could find a transparent fuel (I really don’t know what), it might be possible to use solar heating in the ‘combustion chamber’. Using something similar to a vortex engine, you could keep the chamber walls cool enough to use a transparent material. Instead of injecting fuel for combustion, you could inject light absorbent particles at the top of the chamber. Since we are using our possibly non-existent transparent propellant, only the centre column of propellant/particle mixture will be heated. This is of course highly speculative.

  8. Eric Collins says:

    I’m concerned that using such a large expansion nozzle is likely to waste a lot of the thermal energy by allowing the exhaust gasses to depart in directions which significantly deviate from the desired thrusting direction.

    I also agree with MG that direct heating of the exhaust in the nozzle may not be the most efficient way of heating the gas. You would need to maintain a certain particle density in order to intercept a significant fraction of the incident light and to maintain an effective pressure against the nozzle walls, which is what will ultimately provide the vehicle with thrust.

    It seems that what you propose is similar to the ablative concept that has been put forward in the beamed propulsion community. Would you agree with that comparison?

    I do like the idea of the linear aerospike configuration. The only change I would make is to do most of the heating of the gas in the aerospike thrust chambers rather than in the exhaust plume.

  9. Pete says:

    Maybe one could use regolith as an ablative nozzle surface with which to heat say lunar LOX (it is a good insulator). What regolith is not ablated might become payload. If large quantities of regolith can be cheaply moved into lunar orbit then there are many more options.

    I have also been wondering about using thrust augmentation on a high ISP electric rocket (add a nozzle and inject say LOX into the exhaust). This might greatly increase the thrust to weight ratio, perhaps allowing lunar launch, and might cheaply allow more optimal lower ISP for say Earth-Moon transport systems.

  10. johnhare john hare says:

    I’m trying to wrap my mind around some new possibilities raised by these comments. While I am thinking in terms of wanting an opaque fuel to absorb energy as fast as possible, the idea of a transparent fuel vortex cooling transparent chamber walls with the particles in the center is a new idea for me. Possibly a solid grain in the middle ablating smog?

    I can’t quite yet see how to get the heating inside the chamber with incoherent sunlight. There are a few things stirring in the brain other than Tim’s idea, but none of them have jelled yet. Ablative wasn’t what I had in mind although I can see the similarities in operation.

    Are you suggesting a sun/dirt rocket? I haven’t given a lot of thought to the electric propulsion methods as they all seem to require a heavy power source.

  11. Pete says:

    Yes, a solar regolith engine where the solar mirror ablates the regolith directly, thereby heating a LOX flow. But thinking about it, reheat on the exhaust plume is kind of critical if the solar concentrator is to be low temperature and lightweight (~x5000 concentration probably required for ~2500K). That means to some extent heating the flow directly. Realistically, it will not be getting hot enough via solar concentrators to warrant ablative materials.

    There was a “water jet” rocket that I was thinking about where fine regolith is dispensed down the center of a high pressure/speed oxygen flow. Maybe that could serve as a solar collector. But I would be a little concerned about distributing dust in lunar orbit.

    Assuming thin film solar at ~4kW/kg, I have been wondering what ISP could be obtained from an electric motor power supersonic LOX cooled ducted fan (accelerates and heats flow prior to a nozzle), I suspect around 300, though maybe higher. Such an engine might have a reasonable T/W. If a few extra photons could be focused on the solar cells at launch without over heating them (briefly enabling say 10kW/kg), this could perhaps also work for getting on and off the moon.

  12. Stupid question:

    How fast? Propusion concepts like nuclear thermal and VASIMR promise dramatically-reduced transit times to Mars and beyond. For manned missions that is a big plus.

    Am I asking the wrong question in the wrong place, perhaps? The concept in this blog post of yours may work for slow transit times for unmanned payloads.

  13. A_M_Swallow says:

    @Roderick Reilly

    How fast is an unanswerable question because the question is incomplete. The simplified equations that apply are
    F = m a
    v = u + a t

    To get v we would need to know the mass m of the rocket, the force F (also called the thrust) and the burn time t. The mass changing as fuel is burnt makes everything harder. Given sufficient thrust and fuel we can make the rocket go as fast as we want (until relativity kicks in).

  14. johnhare john hare says:

    There are plenty of stupid questions, but this definately isn’t one of them.

    If it is possible to use this concept as intended, then it should be better than either. Please note that I said if. Higher Isp and T/W than nuclear, and much higher T/W than electric anything at probably 40% of that Isp. For acceleration and fast transit, a clear (note IF) winner on transit times.

    Nuclear thermal Isp is limited by the temperature limits of the heat exchanger. Since the hydrogen must be heated by a physical heat exchanger, there are serious limits to the temperatures attainable. If the hydrogen with a molecular mass of 2 could be heated to the same temperature as a stochiometric chemical H2/O2 engine exhaust product with a molecular mass of 18, then it could have an exhaust velocity 3 times as high.* That would be Isp=~1,350. Actually less because the chemical engines run rich which lowers the average molecular mass.

    Anything electric has powerplant mass issues that kill T/W to the point of uselessness for reasonable acceleration and fast trips. IMO of course.

    The nuclear material temperature limits are what got me thinking about this afterheating trick. A laser from dead astern would be an ellegant solution, unless you have to carry it and all it’s gear along.

    *I’m sure you are aware of the exhaust velocity being proportionate to the square root of temperature over molecular mass. Just letting you know that’s where I got my numbers.

  15. Nels Anderson says:

    Just a couple of comments about opacities. Typical lunar dust particles are about 10 microns in diameter or larger. At visible wavelengths ilicate grains this size present a few hundred square centimeters of absorption cross section per gram. If you can pulverize the dust to get it down aboyt half a micron in size, you can get the opacity up to a few *thousand* cm2/g.

    The opacity of hydrogen does increase quite a bit once the hydrogen is ionized, but 1) that means making it pretty damn hot, 2) the opacity is still small (less than 1 cm2 per gram of hydrogen), and 3) the opacity is principally *scattering* rather than the desired absorption. For heating hydrogen directly, I imagine your better off with an RF beam, and you’d probably want to introduce some some magnetic fields and start using plasma effects. It rapidly distances itself from the simple concept you propose.

  16. johnhare john hare says:

    Thanks for the information. The way I see it, if it goes complex, it goes away as there are plenty of other propulsion ways out there without adding more. Only if it can give serious advantages for the dollar do I see pushing on.

    “Ideas are cheap, implementation is hard.” Henry Vanderbuilt

  17. Habitat Hermit says:

    Hmm a “thermospike”/thermal sail? I think the first illustration is the best/simplest one, a fixed structure of reflective material behind the engine which should work just as well in any direction perpendicular/tangential to the Sun and with decreasing efficiency at angles away from the perpendicular. Making the reflective “sail” axially symmetric through 180 degrees only and possible to rotate in the plane perpendicular to the “top to bottom” axis (i.e. around the nozzle) should allow for additional maneuverability.

    As long as the resulting vector is in the wanted direction I don’t think there’s much point in mirroring the setup in any way for the purpose of achieving “balance” as any such should be able to be defined by the shape of the reflective surface and the fuel expansion properties (is this just counter-intuitive or am I wrong? Perhaps think of it as if it was an aerospike engine for use along a hypothetical sharp border between atmosphere and vacuum: there would only be a point in having a properly “tuned” custom “half annular aerospike” that dipped into the atmosphere just like a boat engine).

  18. Eric Collins says:

    It may be possible to achieve the desired effect using an ionized gas and beamed energy in the form of microwave radiation. If you can send the beamed radiation at the cyclotron resonance frequency of the ions and/or the electrons, and then focus that energy on a small magnetic bottle/nozzle, you might just be able to get some of the advantages of VASIMR in a more efficient package. The use of magnetically confined plasma will allow much higher temperatures – and therefore exhaust velocities – while the use of beamed power removes the mass penalty associated with the reactor and cooling systems. Unfortunately, I’m not familiar enough with the design requirements of microwave reflectors/waveguides to be able to say if they would be any more mass efficient than solar reflectors.

  19. johnhare john hare says:

    If you can use any of the controled beam methods, It should be superior in Isp performance to the mirrors idea. It will almost certainly cost more and have a lower T/W ratio. If you can beam from elsewhere though, performance could be incredible.

  20. A_M_Swallow says:

    A space-port can have a power beaming device aimed at the launch pad. A second power transmitter could be build to handle the launch vehicle when it goes over the horizon. This could work on the Moon, and possibly be solar powered.

  21. johnhare john hare says:

    When funds are available for such things, I want to buy the papers from the last beamed propulsion conference. The one before last had at least one presentation on Lunar laser heating LLOX for Lunar launch. I’m sure that there are many other interesting applications there.

Leave a Reply

Your email address will not be published. Required fields are marked *