PDA

View Full Version : Is a Thermite Core Hydrogen Thermal Rocket feasible?



Ara Pacis
2008-Oct-07, 10:33 PM
It's an idea I cobbled together the other night and I'm curious to know if this would be workable, perhaps for an upper stage or space engine.

Basically, it would use thermite (Al + FeO formulations), which burns at about 2500 C, to heat the working fluid/propellant, probably hydrogen. I'm not sure how to calculate the Isp or thrust for such a configuration, but looking at thermal rockets (solar thermal and nuclear thermal) that operate at similar temperatures, they have Specific Impulse figures around 900 seconds.

I'm probably missing something, lots of things that make this a poor design, but maybe someone could point them out. I'm trying to teach myself rocketry and it's slow going.

Any suggestions or criticisms.

cjl
2008-Oct-07, 10:35 PM
The problem is that the thermite would be consumable, so the fuel weight would include both the weight of the thermite and the weight of the hydrogen. This would mean that the specific impulse would be rather mediocre, as the thermite would not contribute to the production of gas, but would be rather heavy.

Nowhere Man
2008-Oct-07, 11:59 PM
And the thermite reaction produces molten iron. You'll have to deal with that as well.

Fred

Ara Pacis
2008-Oct-08, 12:38 AM
Yes, those are both true. However, this is essentually true for both solar thermal and nuclear thermal systems. The main difference would be in "fuel" duration. The solar thermal is limited only by the sunlight it can gather. The nuclear thermal is limited by the practical lifetime of it's nuclear fuel. Both are much longer timeframes then thermite, but I'm only considering this for a disposable rocket that only fires for the few minutes it takes for an upper stage to reach orbit, for example.

JustAFriend
2008-Oct-08, 01:32 AM
If I remember right, the 2500degrees of the thermite would be three to five times COLDER than the temperature of a rocket exhaust.

If so, you're not gaining anything using the thermite and you could just be causing problems.....

Jerry
2008-Oct-08, 02:03 AM
The limitations on thermite as fuels has more to do with the physical properties and stability than the specific impulse. If you filled a large rocket case with thermite and lit the fuse, when the case expanded the thermite would fracture and crack; the flame front propagate in the cracks and there goes your motor. Many solid rocket fuels contain aluminun and very limited amounts of iron oxide.

cjl
2008-Oct-08, 06:07 AM
Yes, those are both true. However, this is essentually true for both solar thermal and nuclear thermal systems. The main difference would be in "fuel" duration. The solar thermal is limited only by the sunlight it can gather. The nuclear thermal is limited by the practical lifetime of it's nuclear fuel. Both are much longer timeframes then thermite, but I'm only considering this for a disposable rocket that only fires for the few minutes it takes for an upper stage to reach orbit, for example.
The thermite is a consumable that must be carried with the rocket though - solar doesn't need to carry anything comparable, and nuclear, though it does need to carry fuel, only needs a tiny amount. The large weight of thermite required eliminates any advantage it would have over standard rockets. As mentioned before, LH2 and LOX burn hotter than thermite.

Ara Pacis
2008-Oct-08, 08:37 AM
I only mention the solar rocket for thermal comparison, but it wouldn't fit the mission profile of a disposable upper stage to orbit. However, some solar thermal rocket designs (solid particle rotating-bed) would also have additional unexpended mass as part of a heat exchanger. A nuclear thermal rocket also has additional unexpended mass in the form of nuclear fuel and radiation shielding.

If you believe Wikipedia (IYBW) the Space Shuttle Main Engines operate at 3300 C. This is higher than the thermite, which would be around 2500 C. IYBW, the solid core nuclear thermal rocket designs had exhaust temperatures around 2400 C. IYBW and if I understand correctly, the advantage to the hydrogen propellant designs is due to increased translation efficiency in the specific exhaust products: hydrogen versus water steam.

The design of a thermite core would be key, but as it is meant to be a fuel instead of a propellant, it probably wouldn't resemble a solid rocket motor. It would probably have more in common with a solid-core nuclear thermal rocket and be designed with a lots of propellant channels and have a large surface area, or perhaps it would resemble a rotating-bed or a lightbulb design. The burn-out mass of the engine might be high, due to the fuel slag, although I wonder if that could be ejected during the burn without a performance penalty.

I'm trying to figure out if it would provide similar performance to a nuclear thermal rocket during it's short mission profile for the same total engine mass, which was about 6800kg for the NERVA specification. The problem is, I don't know what the actual mass of the fuel was or if that figure includes radiation shielding and other masses that wouldn't be needed for a thermite core.

cjl
2008-Oct-08, 06:37 PM
I only mention the solar rocket for thermal comparison, but it wouldn't fit the mission profile of a disposable upper stage to orbit. However, some solar thermal rocket designs (solid particle rotating-bed) would also have additional unexpended mass as part of a heat exchanger. A nuclear thermal rocket also has additional unexpended mass in the form of nuclear fuel and radiation shielding.
Yes, but the difference is that for both the solar and the nuclear ones, that additional mass is not expendable - it remains the same no matter the burn time. Thermite would be expendable - the longer the burn required, the more thermite required, therefore it is part of the fuel, and also factors into the specific impulse calculations (killing any chance at a decent specific impulse).

nauthiz
2008-Oct-08, 07:07 PM
The Space Shuttle's SRBs use a propellant that's a mixture of aluminum, iron oxide, and ammonium perchlorate, so it's not entirely dissimilar from thermite.

I wasn't able to find complete data, but it looks like one advantage of the SSRB's fuel is that the propellant is much denser than liquid hydrogen (looks like maybe 25x as dense), so the gas expansion rate will be much greater. Also, I'm guessing that with straight thermite there might be problems with gobs of slag compromising the exhaust flow.

Ara Pacis
2008-Oct-08, 08:06 PM
cjl, the fissionables are expendable too, just over a longer lifetime. I think many plans call for reuse of the nuclear engine. I'm not thinking of thermite as an omni-fuel, but instead for a limited mission profile: to achieve a delta-v of around 6000 m/s in vacuum/near-vacuum. I don't know if it will work. I came here to get help finding out.

From what I can see, thermite has fairly low energy density by mass at about 4 MJ/kg, but higher than LH by volume at about 20 KJ/L. I'm not sure how to convert that into effective exhaust velocity, Isp, or thrust. I should have taken Physics my senior year instead of advanced chemistry.

nauthiz, the slag problem is why the core design is important. It could probably be overcome. The question is if the system is useful enough to bother with.

ravens_cry
2008-Oct-08, 08:25 PM
You can make a rocket that uses Salami and NO2, and it could theoretically get you to the moon. That doesn't mean it is practical.

cjameshuff
2008-Oct-08, 09:37 PM
The Space Shuttle's SRBs use a propellant that's a mixture of aluminum, iron oxide, and ammonium perchlorate, so it's not entirely dissimilar from thermite.

It's not at all like thermite, beyond being a solid mixture that can sustain a redox reaction. The aluminum is a fuel, yes, but the iron oxide is a catalyst to promote even, fast, and complete burning, much of the fuel is a synthetic rubber compound, and the exhaust is gas and small particulates.

Thermite is heavy and its combustion products are inconvenient. It's a terrible choice for a rocket fuel, no matter how spectacular it looks when you light a pile of it. You'd be much better off carrying a some liquid oxygen along and burning a portion of the hydrogen to heat the rest. I suspect you'll find the optimum mix is pretty close to what LOX/H2 rocket engines already use. And when you look at the tanks needed to carry the H2, sacrificing a bit of Isp and using a hydrocarbon fuel starts looking real attractive.

nauthiz
2008-Oct-08, 10:20 PM
It's not at all like thermite, beyond being a solid mixture that can sustain a redox reaction. The aluminum is a fuel, yes, but the iron oxide is a catalyst to promote even, fast, and complete burning, much of the fuel is a synthetic rubber compound, and the exhaust is gas and small particulates.
Ah, gotcha. I thought the rubber compound was just a binder and the reaction was aluminum as the fuel and perchlorate as the primary oxidizer (rather than a metal oxide as in thermite). So I was thinking of the difference as being more along the lines of hydrogen/oxygen vs. hydrogen/fluorine.

Ara Pacis
2008-Oct-08, 10:51 PM
cjameshuff, I was looking at LOX/H2 and LOX/RP-1 and nuclear, though I'd prefer to avoid nuclear for several reasons. I was just curious how a thermite core might perform. I'm looking for something fairly compact (for aerodynamics reasons) with good performance. From what I read, the performance from the higher Isp of H2 (~450 seconds) versus RP-1 (~350 seconds) is neutralized or worse by the heavier tankage (10% for H2 versus 1% for RP-1).

I was thinking about nuclear thermal and realized that thermite might provide a similar temperature profile, but for a much shorter time. the question is will it provide it long enough to be useful for the mission profile specified. Moreover, will the thermite continue to burn if H2 is sucking away all its combustion energy (a problem fissionables don't have). I know that the energy density of thermite is ~4MJ/kg and I'll assume half the mass of a NERVA design which would be ~3400kg as the fuel load of thermite for a total energy potential of ~13,600 megajoules (not including inefficiency losses). I'm hoping someone can show me how to convert that to thrust and Isp since I don't know how. I guess I'm looking for a physics lesson.

I'm trying to design a comprehensive space infrastructure for both a fictional setting and non-fictional goalsetting. This is just a tangent from that "research".

cjl
2008-Oct-09, 02:41 AM
cjl, the fissionables are expendable too, just over a longer lifetime. I think many plans call for reuse of the nuclear engine. I'm not thinking of thermite as an omni-fuel, but instead for a limited mission profile: to achieve a delta-v of around 6000 m/s in vacuum/near-vacuum. I don't know if it will work. I came here to get help finding out.
The longer lifetime is what gives it a high ISP though. The high fuel mass of thermite for even a fairly short burn completely kills its effectiveness.

Oh, and as for the SRBs, they do use the aluminum as a primary fuel (along with the rubber binder), and ammonium perchlorate as the sole oxidizer. This is vastly different from thermite though, as this reaction has significant gaseous products (perfect for a rocket motor), whereas the thermite has mostly liquid and solid byproducts.

cjameshuff
2008-Oct-09, 03:37 AM
I'm trying to design a comprehensive space infrastructure for both a fictional setting and non-fictional goalsetting. This is just a tangent from that "research".

I can only see thermite being used as a jury-rigged propellant, used by someone who just doesn't have anything better. Unlike a long-lived nuclear thermal rocket, it will have a hard time moving a mass of propellant at all comparable to the mass of thermite fuel at any reasonable exhaust velocity. Big, heavy, mostly thermite by mass, low thrust to weight, and very short duration.

You would be better off doing something like burning a slurry of aluminum powder with a hydrocarbon as another fuel and as working fluid...the approach used in those solid fuel boosters. (Aluminum burns hot, but hot alumina powder isn't very useful for propelling anything...but it does heat the CO2 and H20 from the burning synthetic rubber material, which provides the working fluid as well as being a fuel.)

One possible approach would be a hybrid using aluminum powder in a synthetic rubber or high melting point paraffin, with LOX as the oxidizer. The fuel modules would be safe to have around, unlike solid rocket fuel grains, and the oxygen's a waste product of aluminum production. Might be a good way to get better use out of imported hydrocarbons. And unlike a thermite engine, it isn't carrying most of its weight in useless iron, and can be shut down once started.

Metricyard
2008-Oct-09, 04:22 AM
One possible approach would be a hybrid using aluminum powder in a synthetic rubber or high melting point paraffin, with LOX as the oxidizer. The fuel modules would be safe to have around, unlike solid rocket fuel grains, and the oxygen's a waste product of aluminum production. Might be a good way to get better use out of imported hydrocarbons. And unlike a thermite engine, it isn't carrying most of its weight in useless iron, and can be shut down once started.


Spaceship One used something similar (http://science.howstuffworks.com/spaceshipone5.htm).

SpaceShipOne is propelled by a mixture of hydroxy-terminated polybutadiene (tire rubber) and nitrous oxide (laughing gas). The rubber acts as the fuel and the laughing gas as the oxidizer.

As far as using thermite, can't see the use. There are many ways to get the results you need, as far as impulse is concerned.

Ara Pacis
2008-Oct-09, 07:52 AM
Well, I was wondering if it might be useful as a non-hydrocarbon rocket propulsion system that might otherwise be relatively accessible in low-g solar system objects. Aluminum and oxygen could be mined on the moon while iron could be mined in asteroids, NEOs and Mercury (if unavailable on the moon). It might be cheaper and simpler to mine in bulk than import hydrocarbon rocket fuel from earth or synthesize from Mars or Venus. The propellant could be volatile ices in nearby space objects or perhaps hydrogen captured from the solar wind directly or on lunar/mercurian surface regolith if possible.

Would this be better than the Al-Ox rockets that might be made from lunar regolith?

Warren Platts
2008-Oct-09, 10:05 AM
(Aluminum burns hot, but hot alumina powder isn't very useful for propelling anything.LOX/Al Isp = 283s

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19920006805_1992006805.pdf

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19920010595_1992010595.pdf

How to get Al:
http://www.moonminer.com/Lunar_Aluminum.html

As for thermal rockets, I'd use an Fe/Oxygen thermal lance (http://en.wikipedia.org/wiki/Thermal_lance) to boil LOX or liquid nitrogen. T = 7,000 - 8,000o

PraedSt
2008-Oct-09, 10:37 AM
Warren Platts Thanks for the moonminer.com link; hadn't come across the site before.
Even more off topic, your avatar scares me! Who or what is it??

Warren Platts
2008-Oct-09, 03:45 PM
Warren Platts Thanks for the moonminer.com link; hadn't come across the site before.
Even more off topic, your avatar scares me! Who or what is it??It's me! My mother was abducted by aliens. Thankfully, she did the right thing and decided to carry the baby to term. . . .

ETA: Here's an old article from the Moon Miner's Manifesto on Al/LOX rockets (http://www.asi.org/adb/06/09/03/02/095/al-o-propellants.html).

PraedSt
2008-Oct-09, 04:19 PM
It's me! My mother was abducted by aliens...

:)
Thankfully I've just remembered a scarier one: Robinson's.
Now I'll quietly leave before I'm arrested for hijacking...

cjameshuff
2008-Oct-09, 04:52 PM
As for thermal rockets, I'd use an Fe/Oxygen thermal lance (http://en.wikipedia.org/wiki/Thermal_lance) to boil LOX or liquid nitrogen. T = 7,000 - 8,000o

Note that thermite involves oxidizing aluminum with sufficient energy output to reverse this reaction and still be rather strongly exothermic, and that iron is considerably heavier than aluminum...a bit over twice the mass per mole. The iron thermic lance rocket, while an interesting concept, might well be even lower performance than a thermite rocket, and would certainly be much worse than an aluminum rocket. It's not impossible the easy availability of asteroid iron would make it practical for small delta v rockets, but I feel other options would work much better.

Warren Platts
2008-Oct-09, 05:07 PM
The lunar regolith is loaded with free iron=--not oxidized--that's ready to be melted into wires. All you need is magnet to harvest it. Extracting aluminum from anorthite, unfortunately, is not a trivial task.

As for the Isp of a iron thermic lance thermal rocket using LOX for your reaction mass, I haven't seen an hard calculations or experiments. But basically, it's an oxidizer rich hybrid rocket, and running rich oxidants can improve the Isp over what would be expected with a regular Fe/LOX hybrid rocket. The extra weight of iron might be worth being able to heat up the reaction mass to 10,000oF.

cjameshuff
2008-Oct-09, 05:07 PM
Here's an idea: boron-oxygen thermal rocket. Boron will also burn in an oxygen atmosphere, and the reaction product is liquid at moderately high temperatures (480 C), allowing good heat exchange and easier handling. It is rare, so you wouldn't want to exhaust it, but molten boron oxide can be cast into glassy ingots (pre-heating reaction mass in the process) that could easily be recycled. It's also quite light weight, and can be formed into fibers that would have a dual use as fuel (when exposed to high pressure, high temperature oxygen) and as structural material (boron fiber composites are quite strong and lightweight).

However, I doubt even this would perform sufficiently well to compete against the alternatives.

Ara Pacis
2008-Oct-09, 07:15 PM
I think there are some rocket fuels that were developed to use boron to some extent, not to mention the Zip fuel craze of the 50s. I recall reading an article about using boron in a turbine for automobiles. You'd probably want to use a turbine/rotating bed design for a rocket too.

The iron lance idea is interesting. I'm wondering what delta-v or mission profiles might make use of these types of rockets where low cost and small size are good but high weight isn't so bad. How about using them to power a lunar intercity rocket rail. With sufficient solar energy, thermal and electric, the fuels could be reprocessed, right?

Warren Platts
2008-Oct-09, 07:27 PM
The iron lance idea is interesting. I'm wondering what delta-v or mission profiles might make use of these types of rockets where low cost and small size are good but high weight isn't so bad. How about using them to power a lunar intercity rocket rail.The idea came out of a thread on Al/LOX hybrid rockets at NASAspaceflight.com (http://forum.nasaspaceflight.com/index.php?topic=14380.0) where it was pointed out that widespread use of aluminum is only about 100 years old because of the difficulty in smelting it. Unlike iron ore that can easily be smelted using medieval technology. So the idea was to think of the quickest, cheapest, dirtiest, lunar ISRU rocket.

With sufficient solar energy, thermal and electric, the fuels could be reprocessed, right?I'm pretty sure that once rocket fuel is used up, it's used up. . . . :sad:

Ara Pacis
2008-Oct-09, 07:55 PM
Only if it's reaction mass too. If you maintain the core in some fashion, while only ejecting propellant, then it could be reprocessed, I think.

Warren Platts
2008-Oct-09, 08:01 PM
Only if it's reaction mass too. If you maintain the core in some fashion, while only ejecting propellant, then it could be reprocessed, I think.OK, I see what you're saying. One could reprocess the spent fuel from a nuclear thermal rocket, for example. With the iron thermic lance thermal rocket, though, that would be too much trouble.

ravens_cry
2008-Oct-09, 08:06 PM
Well, with a solar thermal rocket, the nuclear reactor is external. It can't exactly be reprocessed, but it will last long enough it won't likely matter.

cjameshuff
2008-Oct-09, 08:24 PM
I think there are some rocket fuels that were developed to use boron to some extent, not to mention the Zip fuel craze of the 50s. I recall reading an article about using boron in a turbine for automobiles. You'd probably want to use a turbine/rotating bed design for a rocket too.

That article was actually the inspiration for my suggestion. And yeah, boranes have been considered for rocket fuels, but they would consume huge amounts of boron, which isn't *that* common. A thermal rocket like that would be able to retain the boron oxide for recycling.

Here's the article you mention (I assume it's the same one...haven't seen any other discussion of the idea):
http://www.eagle.ca/~gcowan/boron_blast.html



The iron lance idea is interesting. I'm wondering what delta-v or mission profiles might make use of these types of rockets where low cost and small size are good but high weight isn't so bad. How about using them to power a lunar intercity rocket rail. With sufficient solar energy, thermal and electric, the fuels could be reprocessed, right?

Using rockets for something that could use highly efficient regenerative motors or maglev instead, when there's demand for them in applications with no alternative...sounds incredibly wasteful.

As for recycling, the only thing that would make iron-oxygen rockets even remotely practical is the large amount of metallic iron available on the moon and many asteroids. If you can establish large scale industry for smelting iron from iron oxide, you would be better off doing the same for aluminum and using that as fuel. And you almost need such a thing anyway, to get the oxygen.

However, one thing I've considered using thermic lances for is volatiles extraction. Drill a hole in a stony asteroid, blast the bottom to form cracks, and apply lots of heat to get CO2, water, possibly some nitrogen...if there's a decent metallic iron content, inject oxygen into the rock itself to ignite it. Methane and ammonia would probably be oxidized if there's any present, but the end products would still be of value.

Ara Pacis
2008-Oct-09, 09:10 PM
Some solar thermal designs use carbide pebbles or sand as a heat transfer medium.

Well, it would depend on the feasibility of comprehensive electrification of the rail. Depending on how much electricity is available for power, how durable the rails and/or power lines are in that environment and how widespread is the rail network. Solar power isn't available at night and transmission from the day side might suffer transmission losses and continual degrdation of the transmission lines, assuming there are enough bases or electrical stations scattered around the moon. Other sources of power (nuclear or flywheel storage) might power sections of the rail, if there is enough surplus energy. The day-night thermal cycle might cause problems with a continuous rail and electrical electrical connections between them, which might be a simple maintenance issue, but would prevent the line from being used for some period of time. If the rail network is more widespread and used often, it could be a significant drain on power.

I do like the idea of an electrified lunar rail for short distances and in lunar vacuum subways. But it will be a long time before there is an expansive subway system (though building a covered railway might result in similar performance), so it might be best for trains to be self-locomotive. Maybe it would be better to use thermite and a working fluid in a closed turbine system though.

cjameshuff
2008-Oct-09, 10:11 PM
Well, it would depend on the feasibility of comprehensive electrification of the rail.

No it wouldn't. There are numerous forms of onboard power that are more feasible and less wasteful than thermite rockets. Flywheels, small reactors, or hydrogen, aluminum, or iron fuel cells, or some rechargeable battery (like the vibration and thermal shock tolerant nickel-iron battery) could all provide the necessary power. And if it's a maglev rail, losses may well be low enough that it only needs power at the beginning, and returns most of that energy on arrival.

Ara Pacis
2008-Oct-10, 06:10 AM
No it wouldn't. There are numerous forms of onboard power that are more feasible and less wasteful than thermite rockets. Flywheels, small reactors, or hydrogen, aluminum, or iron fuel cells, or some rechargeable battery (like the vibration and thermal shock tolerant nickel-iron battery) could all provide the necessary power. And if it's a maglev rail, losses may well be low enough that it only needs power at the beginning, and returns most of that energy on arrival.

For some reason I thought Maglev would need an electrified track. Double checking, it looks like you could get by with permanent magnets. Is there a fair amount of neodymium and boron on the Moon?

cjameshuff
2008-Oct-10, 01:40 PM
For some reason I thought Maglev would need an electrified track. Double checking, it looks like you could get by with permanent magnets. Is there a fair amount of neodymium and boron on the Moon?

For a system based on Halbach arrays, you can use permanent magnets, though samarium cobalt would be better due to the temperature extremes on the moon. The magnets are only needed on the vehicles, however, the track material can be simple aluminum or copper. Other methods use electromagnets with iron rails or unpowered aluminum/copper rails. Linear motor/generators at the ends of the track accelerate the train or decelerate it and recover the remaining energy.

To partially answer your question...asteroids contain relatively large amounts of rare earth metals, but it may not be concentrated in easily gathered minerals. On Earth, they are concentrated in minerals associated with granite and various igneous rocks. The moon has had the geological activity needed to form useful deposits, or asteroid nickel-iron in the regolith may contain useful amounts.

Maglev also has the advantage of not being chewed up by the abrasive lunar dust.

Ara Pacis
2008-Oct-10, 06:49 PM
Thanks, I learn something new everyday.

DrRocket
2008-Oct-11, 12:04 AM
It's an idea I cobbled together the other night and I'm curious to know if this would be workable, perhaps for an upper stage or space engine.

Basically, it would use thermite (Al + FeO formulations), which burns at about 2500 C, to heat the working fluid/propellant, probably hydrogen. I'm not sure how to calculate the Isp or thrust for such a configuration, but looking at thermal rockets (solar thermal and nuclear thermal) that operate at similar temperatures, they have Specific Impulse figures around 900 seconds.

I'm probably missing something, lots of things that make this a poor design, but maybe someone could point them out. I'm trying to teach myself rocketry and it's slow going.


Any suggestions or criticisms.

Your idea is actually rather close to what goes on in solid rockets for space launch applications. Those formulations contain about 20% aluminum which is there to make the gasses hot. Generally speaking the Isp of a propellant varies like sqrt(temp/molecular wt). The actually formula is a bit more complicated and involves other terms including the ratio of specific heats but the general relationship is correct. The reason for the high Isp of solar thermal (which you won't find in practice) and nuclear thermal rockets, as well as arc jets (which are more practical than solar thermal rockets and the reason that you won't find the solar thermal variety) is that they provide high temperature to a working fluid which is often hydrogen. So they have the advantage of high temperature and low molecular weight. They also have the advantage of having no other expendables, so only the high temperature hydrogen contributes to the thrust. The disadvantage is a rather high overall inert weight of the vehicle, so they have poor mass fraction.

Your thermite idea has consumables in terms of the thermite components, and those are relatively heavy, higher in molecular weight and may be retained as ash in the rocket. Also the temperature of 2500 C, or 2773K is relatively cool. A typical solid rocket has a flame temperature on the order of 3400K or so.

Ara Pacis
2008-Oct-11, 02:00 AM
...So they have the advantage of high temperature and low molecular weight. They also have the advantage of having no other expendables, so only the high temperature hydrogen contributes to the thrust. The disadvantage is a rather high overall inert weight of the vehicle, so they have poor mass fraction.

Your thermite idea has consumables in terms of the thermite components, and those are relatively heavy, higher in molecular weight and may be retained as ash in the rocket. Also the temperature of 2500 C, or 2773K is relatively cool. A typical solid rocket has a flame temperature on the order of 3400K or so.

But this idea also uses hydrogen as a propellant.

cjl
2008-Oct-11, 05:54 AM
That still fails to solve the problems of the large mass of consumables, as well as the relatively low temperature.

Ara Pacis
2008-Oct-11, 05:58 AM
That still fails to solve the problems of the large mass of consumables, as well as the relatively low temperature.

Assume it's the same mass as a nuclear core. The exhaust of a nuclear is less than the thermite, but I don't know what the core temperature is.

ravens_cry
2008-Oct-11, 06:12 AM
http://en.wikipedia.org/wiki/Nuclear_thermal_rocket
Wikipedia has a lot of interesting information, though the usual warnings about Wikipedia accuracy, of course, apply.

Ara Pacis
2008-Oct-11, 06:46 AM
http://en.wikipedia.org/wiki/Nuclear_thermal_rocket
Wikipedia has a lot of interesting information, though the usual warnings about Wikipedia accuracy, of course, apply.

I've been referring to that since before this thread, but I've been unable to find a reference to an actual core temperature. All I see are exhaust temperatures and the fuel element failure temperature. Maybe I keep missing it.

cjl
2008-Oct-11, 07:48 AM
With an exhaust temp of 2600K, the core temp is definitely well above the 2700K of thermite.

Ara Pacis
2008-Oct-11, 11:57 PM
Speculation is good, but actual answers are better. At ~2500 C, the AL-FEO type Thermite reactions are about 100 C hotter than the exhaust of the Kiwi nuclear thermal rocket tests and about 400 C hotter than the Phoebus nuclear thermal rocket tests. Of course, that's apples and oranges, since I don't know what the exhaust temperature of a thermite core rocket would be or what is the core temperature of a solid-core nuclear thermal rocket. However, IYBW, the Uranium Dioxide melting point is 2846 C and Uranium Carbide is even less (2790 C). Thus, the operating temperature must be less than that, and may well need to be lower due to the lower tolerances of non-fuel structural elements.

cjameshuff
2008-Oct-12, 01:49 AM
Speculation is good, but actual answers are better. At ~2500 C, the AL-FEO type Thermite reactions are about 100 C hotter than the exhaust of the Kiwi nuclear thermal rocket tests and about 400 C hotter than the Phoebus nuclear thermal rocket tests. Of course, that's apples and oranges, since I don't know what the exhaust temperature of a thermite core rocket would be or what is the core temperature of a solid-core nuclear thermal rocket. However, IYBW, the Uranium Dioxide melting point is 2846 C and Uranium Carbide is even less (2790 C). Thus, the operating temperature must be less than that, and may well need to be lower due to the lower tolerances of non-fuel structural elements.

Thermite only produces such temperatures very briefly. You won't be able to heat much hydrogen at all to anything comparable to a nuclear thermal rocket's exhaust before it starts to cool down. You also won't be able to control the engine temperature except by pushing hydrogen through it to cool it. It combines the inflexibility of a solid rocket with more complexity than a liquid chemical fuel rocket, with worse performance than any other rocket type ever used.

Yes, it's a very impressive and spectacular reaction. But it's heavy, and does not in reality produce very much energy. Specific impulse is a tradeoff between energy efficiency and fuel efficiency, high temperatures are used to achieve high fuel efficiency at the cost of energy. A low energy density fuel like thermite is exactly the wrong thing to use for a high Isp engine...it could reach the needed temperature, but won't be able to thrust long enough to be of use. It won't even be very good if operated at a much lower exhaust temperature and higher mass flow rate, because it'll still be heavy *and* it'll need a big reaction mass tank. Given the wide variety of alternatives, some of which being able to use the same materials with better results, there is just no reason to use it for rockets.

Ara Pacis
2008-Oct-12, 05:54 AM
This I know, what I need are numbers.

cjameshuff
2008-Oct-12, 07:27 PM
This I know, what I need are numbers.

Several of the reasons I gave do not need and do not benefit from numbers. You can't turn it off or throttle it, like a solid rocket, but you have a complex engine, even more so than a more standard chemical rocket. And the relationships are easy to infer from my statements about specific impulse:

Ek = 0.5*m*v^2
P = m*v

As exhaust velocity goes up, momentum per unit of reaction mass goes up proportionally, while required energy goes up with the square of the exhaust velocity. The whole point of a high exhaust temperature is to get more thrust from a given amount of reaction mass, and to do that, you need a lot of energy. Combining a low energy density power source like thermite with a high exhaust temperature rocket is exactly the wrong thing to do.

But, here's those numbers you insist on. Iron oxide-aluminum thermite has an energy density of about 4 MJ/kg, or 4 kJ/g. (It'd take about 20 grams to heat up a cup of coffee.) Kiwi got an exhaust temperature of 2683 K. Assume that you get all the energy out as exhaust at that temperature.

The energy required to vaporize and heat a given amount of hydrogen:
E = (904 J/g + 28.836 J/g/K*(T - 20 K))*m

4 MJ = (904 J/g + 28.836 J/g/K*(2663 K))*m
4 MJ/(904 J/g + 28.836 J/g/K*(2663 K)) = 51 g

1000 grams of thermite, if you could get all its energy into heating LH2 to 2683 K, would be able to heat 51 grams of propellant. The maximum achievable exhaust velocity, assuming all energy going into rearward motion (which it won't be, since the expanded exhaust will have nonzero temperature), is around 12000 m/s. Using the rocket equation (delta-v = Ve*ln((mv+mr)/mv), Ve being exhaust velocity, mv vehicle mass, mr reaction mass), the engine has a delta v of around 600 m/s. The thermite fuel itself, that is, with no payload or even structure mass, no matter how big or small you make it. It's a lot of delta v for 51 grams of reaction mass per kg of ship mass, but you'll never get more than that, because the best you can do is a rocket made entirely of thermite and a tiny LH2 tank.

Say you instead went for 1355 K exhaust. That's far more realistic, considering how the temperature will drop as the thermite finishes combusting, and works out as requiring 100 g of LH2 propellant. That gets you up to 850 m/s delta-v with no structure, engine, or payload mass. Say 200 grams/kg of thermite of structure and engine, and 500 g of payload per kg of thermite, that gets you about 450 m/s of delta v for an optimistic but reasonably realistic thermite rocket. That's no help in getting off a surface, and not worth the expense of lifting off the surface with some better rocket. You could use a monstrous number of stages or some way to jettison expended thermite cores and replace them with fresh ones, but that's just adding even more complexity.

You can drop the exhaust temperature even lower, carry more fuel, and in theory get higher delta-v, but you need ever-bigger expansion bells to make up for the low temperature, which themselves add mass and bulk. As the peak temperature drops, you get less and less of your energy output as exhaust velocity, and thus less delta-v.

Nuclear engines, on the other hand, have the energy density to thrust as long as you have fuel as far as trips within the solar system are concerned. They do have a cap on delta v, but it's much, much higher, and a craft using one of these engines would be far out of the solar system by the time it was reached during a single burn.

Ara Pacis
2008-Oct-12, 09:39 PM
Thanks, that's the kinda help I wanted. Now I just have to make sure I understand it all. But I'm not sure how to read this part correctly: Say 200 grams/kg of thermite of structure and engine, and 500 g of payload per kg of thermite, that gets you about 450 m/s of delta v for an optimistic but reasonably realistic thermite rocket. Does that mean a total of 700g of rocket and payload (plus 100g LH2), so 800g to 1000g of thermite for a delta-v of 450 m/s? That would mean the fuel fraction is .06 and the payload fraction would be around 28%, if I work that out correctly. That would be desirable, right? Maybe I'm missing something.

I'm not comparing it to nuclear rockets over all, but just using their performance profiles as a baseline estimate for the duration of the propellant for a single burn in a thermal core type of engine as opposed to the chemical combustion of fuel as propellant.

cjameshuff
2008-Oct-12, 11:20 PM
Thanks, that's the kinda help I wanted. Now I just have to make sure I understand it all. But I'm not sure how to read this part correctly: Say 200 grams/kg of thermite of structure and engine, and 500 g of payload per kg of thermite, that gets you about 450 m/s of delta v for an optimistic but reasonably realistic thermite rocket. Does that mean a total of 700g of rocket and payload (plus 100g LH2), so 800g to 1000g of thermite for a delta-v of 450 m/s? That would mean the fuel fraction is .06 and the payload fraction would be around 28%, if I work that out correctly.

1000 g thermite, 100 g LH2, and 700 g of rocket and payload. The payload was rather arbitrarily chosen, I was assuming you wouldn't want to use much more than twice the payload mass in thermite simply due to cost, and because the rocket is so dismally impractical it doesn't make that much of a difference either way.



That would be desirable, right? Maybe I'm missing something.

...I don't know how you can still see anything about this design as desirable. A low fuel fraction is only good if you can still get a good delta-v. Given the same delta-v, lower fuel fraction is better. A maximum stage delta-v of a few hundred m/s is not desirable!

Nuclear thermal engines are also limited by the power source carried along on the rocket, but their maximum delta-v is much higher than any typical required delta-v that they can operate with very low fuel mass fractions in real-world applications, simply scaling up the fuel mass fraction to achieve higher delta-v. Even with the maximum useful fuel fraction and a infinitesimal payload fraction, you won't get more than 800-900 m/s of delta-v. That's terrible, and to get it you have all the complexities of a solid core thermal rocket, and all the disadvantages of a pure solid fuel rocket.

And given the obvious and far superior alternatives, there's just no reason whatsoever to do it. Just put the aluminum in solid hybrid rocket fuel grains and use LOX as the oxidizer, and you'll have a far simpler, more flexible, safer, and better-performing rocket (or several rockets), and plenty of iron left over for other things.



I'm not comparing it to nuclear rockets over all, but just using their performance profiles as a baseline estimate for the duration of the propellant for a single burn in a thermal core type of engine as opposed to the chemical combustion of fuel as propellant.

That...just doesn't make any sense at all. You seem to have fixated on the high temperatures it can achieve, ignoring the fact that the energy needed to keep it up long enough to do something useful as a rocket just isn't there.

Ara Pacis
2008-Oct-13, 03:36 AM
That...just doesn't make any sense at all. You seem to have fixated on the high temperatures it can achieve, ignoring the fact that the energy needed to keep it up long enough to do something useful as a rocket just isn't there.

I didn't know that until I asked. And I realize I was conflating m/s and Isp in your last post, not sure how I did that.

I wonder if this would be useful as an improvised kinetic rocket weapon for space use.

cjameshuff
2008-Oct-13, 05:23 AM
I wonder if this would be useful as an improvised kinetic rocket weapon for space use.

Not a rocket weapon. An improvised thermite rocket will be lucky to reach a speed that would be supersonic here on Earth, if it doesn't just choke and explode in your face. (Building a solid core with good heat transfer to working fluid that will withstand burning thermite inside it is not a trivial exercise.) Even if it works, it just won't get going fast enough to do more than annoy anything with armor.

A hybrid motor with an oxygen tank (lightweight O2 tanks probably being a common fixture in a space setting, LOX likely being easier to get ahold of than thermite) and a solid fuel grain (the main requirement being "something that burns", something that should be fairly easy to satisfy) would be much easier to improvise, and would be far more likely to get good results.

It could be the basis of a steam bomb, or generate gas for a gun, but again, there are other ways to do those things with more easily available materials. A plain old thermite bomb would be a good device for sabotage or denial of use...especially since many aerospace materials are aluminum based, so with an excess of oxidizer in the bulk of the mixture, the item being sabotaged could contribute some fuel to the cause. In a microgravity environment, the reaction might be violent enough to throw large globs around and cause widespread damage, too...burning lots of small holes in a tank or hull, or damaging multiple items of equipment in a room. That and the more constructive welding applications are really the best things you can use thermite for, though.

Ara Pacis
2008-Oct-13, 07:41 AM
Okay, thanks for all the help and hard work. I'm gonna get back to imagining LH2/LOX upper stages for my imaginary aerospace taxi. Or maybe RP-1/LOX if I can figure out which is better... they're coming out about the same using rules-of-thumb.

cjameshuff
2008-Oct-13, 05:44 PM
Okay, thanks for all the help and hard work. I'm gonna get back to imagining LH2/LOX upper stages for my imaginary aerospace taxi. Or maybe RP-1/LOX if I can figure out which is better... they're coming out about the same using rules-of-thumb.

Look at LOX/CH4. Methane gets much of the advantages of hydrogen (being 25% hydrogen by mass), but is far denser and is liquid at temperatures comparable to LOX (boiling point's a little higher at one atmosphere). It's also a valuable material for synthesizing plastics and other chemicals, so there's plenty of good reasons to have the capability for moving large amounts of it around.

Ara Pacis
2008-Oct-14, 10:24 PM
I've heard that CH4 could be useful, but I haven't looked at it too hard. I probably should. I'm still trying to teach myself rocketry and some of what I read in wikipedia isn't clear, such as r (mixture ratio). If r is oxidizer/fuel then LF2 with an r of 4.83 means there is almost 5 times as much LOX as LH2, which doesn't sound right. And when I look at charts for rule-of thumb, I'm not sure if the mass of the oxidizer is included or not.

Maybe I should start a new thread for this new line of questioning.

cjameshuff
2008-Oct-15, 02:10 AM
I've heard that CH4 could be useful, but I haven't looked at it too hard. I probably should. I'm still trying to teach myself rocketry and some of what I read in wikipedia isn't clear, such as r (mixture ratio). If r is oxidizer/fuel then LF2 with an r of 4.83 means there is almost 5 times as much LOX as LH2, which doesn't sound right. And when I look at charts for rule-of thumb, I'm not sure if the mass of the oxidizer is included or not.

Remember that an oxygen atom weighs 16 times as much as a hydrogen atom, and that ratio is by mass. A stoichiometric ratio of O and H would be 1 oxygen atom per 2 hydrogen, 16 to 2, or 8. A ratio of 4.83 means there's almost twice as much hydrogen as can be oxidized, the remainder is just reaction mass. As I understand it, gasses with diatomic molecules like H2 trap less thermal energy in vibrations and rotations as they expand, so you don't need as much of an expansion bell to get it out as linear motion. The high molecular mass of oxygen makes it less useful for capturing rotational and vibrational energy from the H2O molecules in the exhaust...there's just 16 times as many molecules in a kg of H2 than in a kg of O2.

Complete combustion of kerosene, assuming 14 carbon chains, requires a ratio of around 3.5, but again, a larger number of smaller molecules in the exhaust is preferable, so a rich mixture that results in a lot of nicely diatomic CO is used.

Ara Pacis
2008-Oct-15, 04:07 AM
D'oh! I knew I was missing something, but I didn't realize it was something as basic as stoichiometry. It's been 17 years since I had to use moles and the like.

Now if only I can figure out if those delta-v/Ve by Mass ratio charts include oxidizer.