NASA will join a military program to develop nuclear thermal propulsion

None of this will happen quickly. The technology is difficult and unproven.

See full article...

 

[umaxtu]

Weird related note, but somehow the Kalamazoo Air Zoo got their hands on a nozzle for a NERVA engine. It's packed to the gills with sensors and test leads to boot.

IIRC it's kinda tucked in a back corner of the restoration hanger. I wonder if they will be putting out a YouTube video about it in response to this news. They've actually got some pretty great content on their channel.

 

[cerberusTI]

Hydrogen strikes me as having several problems. The first is that it is difficult to store long term cryogenically. The second is that it is low mass, so has to be expelled at very high velocity to produce an appreciable amount of momentum. The ionic propulsion units use heavy metals like mercury for this reason. Third, if the exhaust temperature is high enough the hydrogen will be at least in part coming out as atomic hydrogen. Any energy dissociating molecular hydrogen is wasted, as even if recombination occurs after expulsion, the energy recouped will not go to momentum transfer to the rocket. And atomic hydrogen is very reactive chemically, so may cause all sorts of problems in the engine itself. I would think that an inert gas like helium would be better (except for the storage problem) but neon, or, at the other mass extreme, xenon, would make better propellants.

It would not be wasted energy, although it would take a lot of energy.

Exhaust velocity and the molecular mass of what you are exhausting are related. If you are exhausting atomic hydrogen instead of H2, you halve the mass. The thrust would be the same for more energy so it is a waste from that perspective, but the ISP would be far higher so it would also be much more efficient in propellant use.

 

[cerberusTI]

If you can make (and store) metallic hydrogen, I don't think the nuclear reactor is necessary. There's a paper giving a theoretical max I_SP for a metallic H2 monoprop of 1,700s (at the cost of a 'combustion temperature of 6,000K)

NERVA only managed just under 850s

Similar to my last post, that high ISP is due to the high combustion temperature which can dissociate hydrogen. The source of the hydrogen does not matter so long as you get it hot enough.

You do not need metallic hydrogen for that, it would just be much denser when stored.

 

[DStaal]

Wasn’t there a proposal for using nuclear explosions? Shuffled out the back of a spacecraft to propel it forward?

That's the fabled Project Orion. It's actual safer than it sounds, and would have been great...

There's also the Medusa rocket - throw the bombs out the front of the rocket, and have a sail in front to catch the explosion. Reduces the mass needed significantly, making it even better.

And if you want to go really crazy: Zubrin's Nuclear Salt Water. Don't package the Orion fuel into little bombs you send off and explode behind you, instead keep it as a liquid solution and explode it in the engine itself.

 

[DStaal]

Similar to my last post, that high ISP is due to the high combustion temperature which can dissociate hydrogen. The source of the hydrogen does not matter so long as you get it hot enough.

You do not need metallic hydrogen for that, it would just be much denser when stored.

But the point of metallic hydrogen is that the hydrogen is now both the fuel and the propellant: As it decompresses from metallic form, it releases enough energy on it's own to get it hot enough to compare to nuclear thermal, without the fixed dry mass. (Assuming you can keep it in metallic form until it's needed...)

 

[normally butters]

This is very cool news, despite the convoluted bacronym (the military should've asked NASA to come up with it, they're masters at that by now).

Obviously, there is a lot of research to do, but it's great that it's finally happening.

Nuclear propulsion (thermal or electric) doesn't make a lot if sense for getting in and out of atmospheres, but for pure space transportation, tugs, cargo, etc. it's unbeatable -- since we're already dealing with a radioactive wasteland to begin with.

One way to avoid public nucular-fobia would be to launch the spacecraft with reactor un-fueled and get the fuel elsewhere -- either from the Moon, enriching it in orbit (unlikely, but still an option) or using low-risk nuclear fuel such as Thorium. Not to mention, we already have launched about 5kg of plutonium to Mars on each of the last two rovers (Curiosity & Perseverance).

All of this would require advances to in-space assembly and/or manufacturing, but given the long-term prospects of the research, we'll need all that by then anyway, after all fuel depots and in-orbit refueling are already integral to Artemis.

Yeah nobody is arguing for nuclear propulsion for launch vehicles, but the thing is that nuclear thermal propulsion is limited in what it can do for in-space propulsion as well. It's not particularly useful outside of the context of an Earth departure stage that does one burn to send the spacecraft on its way. It wouldn't make a good space tug and would barely be capable of any complex orbital maneuvering at all.

One big problem is the Xenon Trap. You might recall this from the fantastic final episode of the HBO Chernobyl series. When the uranium-235 is fissioned, one of the fission products is iodine-135, which naturally decays to xenon-135, the strongest known neutron absorber or "poison." But the fission chain reaction also fissions the I135, so the Xe135 doesn't accumulate in the core.

But when the reactor is shut down, it stops fissioning the I135, and the Xe135 neutron poison accumulates in the core until it naturally decays to caesium-135 with a half-life of about 9 hours. Therefore, once a solid-fuel fission reactor is shut down, it must remain shut down for about 3 days until the Xe135 decays sufficiently to allow the reactor to be restarted... safely. Failure to wait for Xenon Trap to decay was one of the negligent acts leading to the Chernobyl disaster..

So in a nuclear thermal propulsion system, when the burn ends and the reactor is shut down -- because there's no non-propulsive way to cool the reactor while it's running -- there can't be another burn within the next few days. At best, it would need the cryo endurance for its tank of liquid hydrogen to loiter on orbit for a few days while waiting for the reactor core to be ready to support a second burn.

DARPA may call this an "Agile Cislunar Operations" solution, but it's not really agile when the restrictions on multi-burn mission profiles are so onerous.

 

[cerberusTI]

And, why specifically hydrogen? Does its behavior as a propellant behavior really outweigh the containment issues?

Anyone know of a chart of nuclear propellants? There's lots for chemical propellants, charting all the usual values, but I've never seen one for nuclear propellants.

It is all down to molecular mass, as your heating comes from the fission fuel.

Atomic hydrogen is the most efficient, but it requires extreme temperatures.

H2 is next, with two protons.

Anything else will basically scale to the molecular weight, so much worse. Water could be used as it is dense and abundant, and not terrible on ISP (although having an oxygen atom at mass 16 is a big drag on ISP compared to just hydrogen at mass 1 or 2).

 

[mynameisntsir]

quite exciting. the heat transfer problems alone seem immensely challenging.

 

[cerberusTI]

But the point of metallic hydrogen is that the hydrogen is now both the fuel and the propellant: As it decompresses from metallic form, it releases enough energy on it's own to get it hot enough to compare to nuclear thermal, without the fixed dry mass. (Assuming you can keep it in metallic form until it's needed...)

Something tells me keeping it in its metallic form will not compare favorably to nuclear in some ways.

In terms of exotic engines, my favorite is the fission fragment rocket. Fission reactions can have enough energy behind them that even a heavy atom is accelerated to nearly the speed of light, and it is really the speed that matters here. Low mass just makes it require less energy.

 

[Hispalensis]

It'll be interesting to see how they intend to keep a bunch of liquid hydrogen both liquid and in a tank for the time required for a mission to Mars (and back again).

You just keep it as a gas and use the additional buoyancy to go up and reach Mars faster /s

 

[destoya]

NTP at its core really isn't overly complex, but there are some significant materials science challenges to overcome first. A NTP reactor is designed to run hotter than conventional ground-based reactor (~2700K) in order to increase efficiency, but which brings serious problems with hydrogen embrittlement and corrosion of existing state-of-the-art Fe- and Ni-based alloys. This corrosion of reactor cladding is the current limiting factor of NTP feasibility and lifetime. I attended a very interesting talk early this week on this subject, seems like the most promising current direction is Mo-W cermet-coated uranium.

Current NASA target is up to 8 restarts (i.e. 1-2 round trips to mars with acceleration and deceleration burns each way) with a 4-8h burn time each way, so this is very much not a reusable system with current materials. Still, if it works out it has obvious advantages versus even the most optimistic Starship mars transfer scheme.

 

[noone.of.import]

Ionic propulsion uses heavy metals despite their high atomic weight because they're easy to ionize. Ideally they'd use lighter propellants as well, but at the moment the advantages of higher ISP are often outweighed by the increase in mass needed to ionize the smaller atoms.

Do you have a link to a good explanation of this trade-off between heavier and lighter elements? It's not obvious to me from physics 101 conservation of momentum, and that's about the extent of my rocket science education.

 

[Peflitydap]

Hopefully the contracts can be spread among all 50 states and numerous house districts. /s

Realistically,
Nuclear is the only known technology for us to be able to colonize this solar system.

Not until we get the improbability drive online.

I thought sufficiently long rail guns were a possibility too.

 

[DStaal]

Something tells me keeping it in its metallic form will not compare favorably to nuclear in some ways.

Oh yeah, definitely. Most of the fuss around Metallic Hydrogen dates back to a couple of papers that calculated that it might be possible to keep it stable in a supercritical state at only moderately cryogenic temperatures and at fairly low pressure. Which would mean storage doesn't require huge amounts of mass on it's own - but it put you in the dangerous position of having a fuel that if you bump wrong goes boom.

Those papers mostly haven't been held up by experiment so far, but the stuff's hard to make anyway so there's some doubt left I believe.

 

[DStaal]

I thought sufficiently long rail guns were a possibility too.

Those are good for getting off this planet. You still need the other half of the delta-V to get someplace.

 

[Veritas super omens]

In what (theoretical) way does one make metallic hydrogen?

Lots and lots of pressure. Like the interior of Jupiter.

 

[DStaal]

Do you have a link to a good explanation of this trade-off between heavier and lighter elements? It's not obvious to me from physics 101 conservation of momentum, and that's about the extent of my rocket science education.

It's a bit long and is a bit more general than just this issue, but in general you can likely find everything you ever wanted to know about rocket engines here:
http://projectrho.com/public_html/rocket/engines.php

 

[Emon]

NERVA is back, baby!

Yeah I’m actually confused by the claims of it being never realized and new technology.

NERVA absolutely worked, they had that burning for hundreds of hours and hundreds of start stop cycles.

I know it’s never been put into space but, the whole point was that it’s actually rather simple. There isn’t even any combustion.

There’s probably more I don’t know but if they did this in the 60s with slide rules it seems a it odd to consider it “out there” in 2023.

I actually met a guy a few years ago, in the Before Times, who worked for SpaceX as a refueler (at a space themed brewery, how odd to find him there!) He worked with hydrazine all day. I asked him about NASA’s “green” hydrazine alternative and to my surprise just shook his head. He said they need to go straight to NTP, it was proven to work decades ago and fear of nUcLeAR is all that’s really stopped us.

Well it seems NASA agrees with him!

 

[Deleted member 485025]

I thought sufficiently long rail guns were a possibility too.

For orbit, a superconducting launch system for cargo seems reasonable.

Not for landing and re launching from another planet or asteroid

 

[Emon]

Thermal nuclear rockets? What could possibly go wrong with that?

Highly combustable fuel in a carriage?? What could possibly go wrong with that?

 

[fenris_uy]

In what (theoretical) way does one make metallic hydrogen?

Grab some hydrogen, apply a lot of pressure.

 

[Emon]

Grab some hydrogen, apply a lot of pressure.

Just mine the core of Jupiter, easy peasy!

 

[ColdWetDog]

You could even use water ice and vaporize it into a surprisingly effective steam rocket.

Ah, that's Arcaspace's long term plan. Aside from the crypto coin, that is.

 

[Ironsights]

I don't understand why we aren't using Epstein Drives. They are pretty well documented in literature.

 

[adam.i]

I thought NASA was involved with DRACO from the beginning.

Hydrogen certainly gets you the best potential specific impulse, but there's an interesting system trade to be made for ammonia - about half the specific impulse, but much lower inert mass due to smaller tanks and no need for cryocoolers.

My favorite story from NERVA how a bird nearly messed up a test series and how President Nixon had to make the call to continue testing:
https://talesfromthenuclearage.wordpress.com/2009/08/start with "wayward traveler" at the bottom then "the decision"

The "Beastie" referred to in the story is the GE "Beetle" Mobile Manipulator. Probably one of the most badass looking machines ever made.

 

[DStaal]

I don't understand why we aren't using Epstein Drives. They are pretty well documented in literature.

Discussed here:
http://www.projectrho.com/public_html/rocket/enginelist3.php
They're likely fairly conventional fusion drives in design, but a lot of cooling is being skipped out on.

 

[JohnDeL]

One way to avoid public nucular-fobia would be to launch the spacecraft with reactor un-fueled and get the fuel elsewhere -- either from the Moon, enriching it in orbit (unlikely, but still an option) or using low-risk nuclear fuel such as Thorium. Not to mention, we already have launched about 5kg of plutonium to Mars on each of the last two rovers (Curiosity & Perseverance).

I admire your optimism. There is a substantial fraction of the public out there that probably thinks Space: 1999 was a documentary and would still get upset about any use of nuclear materials in space.

 

[andygates]

ARCA's secret plan -- EXPOSED!

It's the atomic teakettle, they're just missing the atomic part.

 

[prestonf]

It'll be interesting to see how they intend to keep a bunch of liquid hydrogen both liquid and in a tank for the time required for a mission to Mars (and back again).

NASA has some projects already in progress to address this issue. Definitely a non-trivial issue.
https://www.nasa.gov/directorates/spacetech/solicitations/tipping_points/2020_selections

 

[GlockenspielHero]

That's the fabled Project Orion. It's actual safer than it sounds, and would have been great...

There's also the Medusa rocket - throw the bombs out the front of the rocket, and have a sail in front to catch the explosion. Reduces the mass needed significantly, making it even better.

And if you want to go really crazy: Zubrin's Nuclear Salt Water. Don't package the Orion fuel into little bombs you send off and explode behind you, instead keep it as a liquid solution and explode it in the engine itself.

If you're going completely crazy, why not a closed cycle gas core nuclear thermal (aka nuclear light bulb) rocket?

I love the idea of running a reactor at 22,000C

 

[Tiny Bear Skin Rug]

The article understates the progress made on NERVA. A non-flight-weight rocket was tested several times at Jackass Flats, Nevada, and one was deliberately blown up (so, RED? Rapid Expected Disassemby?) to evaluate the effects of a RUD. One of the issues with the original design was relatively poor thermal transfer of heat to the fuel (although still reached an Isp of, IIRC, about 900sec), so a new design, DUMBO, with smaller, closer-together fuel channels was designed but never tested.

(Caveat: posted while still waiting for my first cup of coffee to brew, I may be misremembering some details.)

Rapid Intended Disassembly. They got RID of it.

 

[Zylon]

Whatever happened to detonating a thermonuclear bomb and riding the shockwave. Would be a blast.
https://en.wikipedia.org/wiki/Project_Orion_(nuclear_propulsion)

Atomic (i.e., fission) bombs, not thermonuclear (i.e., fusion) bombs. For a realistic rocket, each explosion has to be a fraction of a kT; a hydrogen bomb will ruin your day. And you'll need thousands of them. That's actually doable, since each one would be approximately the side of a soda can. None of the people who worked on the project would ever talk about the size of each bomb, of course. But Freeman Dyson's son (who never signed an NDA to obtain a security clearance and could talk about it all he wanted) revealed that interesting tidbit.

 

[uhuznaa]

NERVA XE Prime was a nearly flight-ready NTP engine tested in 1968 and 1969, more than 50 years ago:

"The final test of the series was XE Prime. This engine was 6.9 meters (23 ft) long, 2.59 meters (8 ft 6 in) in diameter, and weighed approximately 18,144 kilograms (40,001 lb). It was designed to produce a nominal thrust of 246,663 newtons (55,452 lbf) with a specific impulse of 710 seconds (7.0 km/s). When the reactor was operating at full power, about 1,140 MW, the chamber temperature was 2,272 K (2,000 °C), chamber pressure was 3,861 kilopascals (560.0 psi), and the flow rate was 35.8 kilograms per second (4,740 lb/min), of which 0.4 kilograms per second (53 lb/min) was diverted into the cooldown system.[1] A series of experiments were carried out between of 4 December 1968 and 11 September 1969, during which the reactor was started 24 times,[89] and ran at full power for 1,680 seconds."

An ISP of more than 700 seconds is something, no doubt. The problem of course always is mission design with a nuclear engine: You have to launch it in an inert and very safe state (which also means you can't static fire it for testing before launch), you won't be using it anywhere for landing, so it's basically limited for a space tug or for a departure stage for heavy interplanetary probes and other uncrewed missions.

If you use it to return to Earth after a mission you are limited to doing a propulsive orbit insertion because you won't land your craft with it and you probably even don't want to do aerobraking or aerocapture because of the risk of strewing highly radioactive debris all over the place in case of a failure.

For me NTP engines are just the next level of Hydrolox over-optimization madness by looking for the best ISP and ignoring all of the problems and limitations you buy into with it.

For some rare mission profiles it may be worth it, yes. For most profiles not so much.

 

[NomadUK]

Running just a little behind schedule ...

 

[andygates]

And you'll need thousands of them. That's actually doable, since each one would be approximately the side of a soda can.

Alexa, define "proliferation nightmare".

 

[alastairmayer]

I admire your optimism. There is a substantial fraction of the public out there that probably thinks Space: 1999 was a documentary and would still get upset about any use of nuclear materials in space.

Wait, you mean we didn't used to have two moons?

 

[Cthel]

Alexa, define "proliferation nightmare".

It gets worse - those tiny nukes would be directional - designed to deliver as much of their energy in a narrow cone as possible (to maximise propulsive efficiency; that cone would be aimed at the pusher plate)