How Can We Have NTR SSTO?

[sanman]

For those who'd say it's absolutely not practical/possible, please at least see this as a thought exercise to make the impossible possible.

If we really, really wanted/needed to do NTR SSTO, what would be the best way to do it?

What could be done to mitigate risks to make flights as safe & reliable possible, and to minimize wider hazards of nuclear contamination in the atmosphere?

Regarding the mass penalty from radiation shielding, could reductions in shield mass be possible for unmanned vehicles?

How could propellants be used for radiation shielding?

How could a nuclear reactor/engine be modularized or ruggedized to help it survive intact, in the event of loss of control over the vehicle or breakup of the vehicle?

Would it be possible to use a highly-enriched nuclear fuel to achieve higher burnup-fraction, for great fuel efficiency?

--------


So I'll confess that my renewed interest in this topic stems from the recent Russian test of its new 9M730 Burevestnik missile, which is purported to use nuclear ramjet propulsion. According to statements from the Russian president, the missile was apparently test-flown for a duration of 14-15 hrs, and apparently monitored by a NATO observation vessel as well as by ground monitoring stations in nearby countries (eg. Norway), with no anomalous radiation levels being reported. The statements also mentioned the reactor using liquid-metal coolant, as per submarine reactors, but being far smaller in size. The flight regime was sub-sonic, but it was claimed that faster versions could one day be possible (hypersonic?) The nuclear fuel was said to be highly enriched to some high fraction below weapons-grade.


Fine, fine, everyone is free to take the statements with a grain of salt, or reject them outright, etc.
But for the sake of discussion, supposing the propulsion system is able to perform as advertised, with its relatively small form-factor.

(Disclaimer: not seeking weapons discussion, just referenced what looks to be recent real world hardware, to serve discussion on how to do SSTO via NTR)


But this then offers the possibility of having a nuclear-powered scramjet-style SSTO that could use the atmosphere for propulsion by staying in the upper atmosphere for as long as possible while ramping up toward orbital velocity. This could allow significant savings in propellant mass and gross takeoff weight. The much greater energy density available from nuclear could help to overcome the tyranny of the rocket equation.

I was thinking that scramjet-style HOTOL would be better than VTVL nuclear rocket, because the flight profile is lower-G and less stressful aerodynamically, which helps safety.
Scramjet-style trajectory however means more heat buildup and thermal issues from frictional heating, due to staying in the atmosphere for longer.
Was also thinking that densified LCH4/LOX (less coking) could be our onboard propellant for getting up to ramjet speed, and later also providing the final boost to orbital velocity.

---


What are the key considerations that would make/break the NTR SSTO, from the technical and overall practical standpoints?

[Starship Trooper]

Back in the early 90s, when SSTO (Delta Clipper, Venture Star) was the aspiration, reusability was part of the package.  Now 30 years later, we have achieved booster reuse with Falcon 9 and Super Heavy, (with a landed New Glenn booster awaiting reuse).  Cost effective upper stage reusability is still an unfulfilled aspiration, but booster reusability has already changed the game.

NTR SSTOs would only make sense if they were fully reusable, and there is the rub.  No one is going to be happy about frequent reentry of NTR SSTOs until safe and reliable reentry of conventional upper stages is a proven and mature technology.  In other words, NTR SSTOs can't even be considered until we have at least a couple decades of routine reliable upper stage reuse experience behind us.  (The Shuttle experience is not encouraging)

[sanman]

Back in the early 90s, when SSTO (Delta Clipper, Venture Star) was the aspiration, reusability was part of the package.  Now 30 years later, we have achieved booster reuse with Falcon 9 and Super Heavy, (with a landed New Glenn booster awaiting reuse).  Cost effective upper stage reusability is still an unfulfilled aspiration, but booster reusability has already changed the game.

Alright, but DC-X & X-33 were chemical propulsion, which imposes excruciatingly tight margins for doing SSTO.
It's not that SSTO is inherently less reliable than multi-stage to orbit, but rather that it's too difficult to achieve with chemicals.
If we were purely comparing nuclear SSTO vs nuclear multi-stage-to-orbit, then you & everyone would probably agree that SSTO helps KISS.  (aka. "best stage is no stage")

NTR SSTOs would only make sense if they were fully reusable, and there is the rub.  No one is going to be happy about frequent reentry of NTR SSTOs until safe and reliable reentry of conventional upper stages is a proven and mature technology.  In other words, NTR SSTOs can't even be considered until we have at least a couple decades of routine reliable upper stage reuse experience behind us.  (The Shuttle experience is not encouraging)

But let's say we're planning for that day when there's been a couple of decades routine reliable chemical upper stage re-use, etc. Things seem to be on track for eventually achieving that.

We want nuclear SSTO not because it's cool/sexy, but because we want to free ourselves from Tyranny of Rocket Equation.
We want to be able to scale to launch vehicles with larger payload mass & mass fraction, to enable better economies of scale and more mission flexibility.
We want to be able to send ships to Mars without having to send up 4 other orbital re-fueling flights just to support the main flight.
We want power/energy to spare, so that we won't have to fret over every gram and kilogram being flown.

We don't know how to do fusion yet, so we want to go with fission, accepting its known issues/non-idealities.

[redneck]

The problem being that you're not free of the rocket equation. And you're stuck with hydrogen for any practical NTR. Even with a feasible NTR, the takeoff mass will be over 60% hydrogen* with attendant tank mass and thermal issues.

*Unless you can somehow run a lot more heat than anyone expects since "normal" NTRs are seriously limited in allowable temperatures.

[Paul451]

Alright, but DC-X & X-33 were chemical propulsion, which imposes excruciatingly tight margins for doing SSTO.

Not as tight as NTRs.

Chemical rockets have thrust/weight ratios upwards of 70:1 (Merlin is closer to 200:1). NTRs are in the theoretical range of 7:1, and more likely 4 or 5:1. Add shielding mass (even for uncrewed rockets) and it's worse. Gravity losses kills NTRs as a first stage. Super-kills it for an SSTO.

NTR would be struggling to be useful as a second stage. AIUI, proposals were to use them for a third stage; ie, once they've achieved minimal orbital velocity and you can switch to high efficiency, low thrust engines for long BLEO cruise phase.

It might, perhaps, be possible to use NTR as a launch vehicle from the moon. But I suspect even Mars would be too much.

Jet engines (and hence air-breathing NTRs serving the same role in a doomsday missile) can have much lower thrust/weight, because wings do the work of fighting gravity, the engines mostly only need to fight drag. [For example, the stupidly powerful F119 from the F-22 has a thrust/weight of around 8:1, giving the whole jet a thrust/weight slightly above 1:1. Modern turbofan engines are around 5 or 6:1.]

[sanman]

Thanks for nice responses

The problem being that you're not free of the rocket equation. And you're stuck with hydrogen for any practical NTR. Even with a feasible NTR, the takeoff mass will be over 60% hydrogen* with attendant tank mass and thermal issues.

*Unless you can somehow run a lot more heat than anyone expects since "normal" NTRs are seriously limited in allowable temperatures.

But hydrogen only seems to be favored for Isp reasons. For SSTO we'd want to balance that against thrust, which methane is better on.
If you're referring to neutron radiation, it doesn't care about what chemical form that hydrogen is in, as long as it has the necessary density. LCH4 has plenty of it, and is easier to densify than LH2.

The thing is that we'd be trying to achieve most of our deltaV via ramjet propulsion, thru nuclear heating of airstream.
The onboard propellant is just for the initial and final phases of our ascent, while everything in between would be nuclear-ramjet.

A main challenge however is how to achieve nuclear-ramjet heating at high mach numbers.
Maybe that's where we'd have to start gradually using onboard propellant mixed into airstream at higher mach flows.

Alright, but DC-X & X-33 were chemical propulsion, which imposes excruciatingly tight margins for doing SSTO.

Not as tight as NTRs.

Chemical rockets have thrust/weight ratios upwards of 70:1 (Merlin is closer to 200:1). NTRs are in the theoretical range of 7:1, and more likely 4 or 5:1. Add shielding mass (even for uncrewed rockets) and it's worse. Gravity losses kills NTRs as a first stage. Super-kills it for an SSTO.

NTR would be struggling to be useful as a second stage. AIUI, proposals were to use them for a third stage; ie, once they've achieved minimal orbital velocity and you can switch to high efficiency, low thrust engines for long BLEO cruise phase.

It might, perhaps, be possible to use NTR as a launch vehicle from the moon. But I suspect even Mars would be too much.

Jet engines (and hence air-breathing NTRs serving the same role in a doomsday missile) can have much lower thrust/weight, because wings do the work of fighting gravity, the engines mostly only need to fight drag. [For example, the stupidly powerful F119 from the F-22 has a thrust/weight of around 8:1, giving the whole jet a thrust/weight slightly above 1:1. Modern turbofan engines are around 5 or 6:1.]

Thanks for this. So thrust is the limitation, due to the problem of working fluid and power transfer to the exhaust stream.

But can't we get away with less thrust if it's an air-breathing ascent trajectory making use of lift from wings?
The fact that Burevestnik could fly for 14-15 hours could allow for a more gradual ascent, while exploiting lift to offset gravity losses. Re-entry could be the reverse of that trajectory. But then there's the issue of heat buildup.

Alternatively, could we elevate thrust using nuclear-chemical hybrid propulsion?
Perhaps something like a "nuclear pre-burner" that heats up reactant flowstreams before they go to main combustion chamber?

It's mind-bobbling that we're TRAPPED in the world of chemicals and their energy limitations. There's got to be some better way.

[Starship Trooper]

It's mind-bobbling that we're TRAPPED in the world of chemicals and their energy limitations. There's got to be some better way.

We are not trapped, there are other potential technologies limited only by imagination and capital investment.

For one, giving your vehicle/payload a quick start with a mass driver is doable, a Chinese company is working on that now.

For another, beamed power from the ground or orbit is another option, allowing use of a monopropellant.

While slightly dated now, the 1991 book by Mallove & Matloff, The Starflight Handbook was a good review of all sorts of propulsion concepts

[Vultur]

Thanks for nice responses

The problem being that you're not free of the rocket equation. And you're stuck with hydrogen for any practical NTR. Even with a feasible NTR, the takeoff mass will be over 60% hydrogen* with attendant tank mass and thermal issues.

*Unless you can somehow run a lot more heat than anyone expects since "normal" NTRs are seriously limited in allowable temperatures.

But hydrogen only seems to be favored for Isp reasons. For SSTO we'd want to balance that against thrust, which methane is better on.

If you use anything other than hydrogen, your Isp is not good enough to *remotely* make up for the bad mass ratio introduced by the heavy reactor+ shielding.

If I wanted to make an SSTO, I'd focus on really low mass fractions not Isp. An expendable SSTO is IMO well within current technology (I believe a tank stretched F9 first stage would manage it, especially with landing legs deleted; F9 first stage is already supposed to be 95.5% propellant fraction or about mass ratio 22, and a tank stretch would improve that). I think a reusable chemical SSTO is probably possible today if you use tricks like Starship's tower catch... Just not economically practical. It'd be huge for its payload size.

It's mind-bobbling that we're TRAPPED in the world of chemicals and their energy limitations. There's got to be some better way.

There are other options, solid-core NTR is just a particularly bad one. (This is sad, because NERVA was super cool and I used to really hope it'd be revived. But I think the practicalities just don't work.)

Most of the other ground to orbit options require a lot of very expensive ground infrastructure, like laser launch, SpinLaunch, advanced gun launch, etc.

Also, "energy limitations" for chemical rockets aren't really the issue. The fuel cost to orbit is quite tiny. It's bespoke hardware and huge staff costs that make launch expensive. Thus, going to nuclear (more expensive hardware, more regulatory costs & thus more staff needed) doesn't help.

But, in the spirit of the thread, if someone wanted to build an NTR SSTO regardless of practicality (and had total exemption from any environmental or safety requirements) the only real way to do it would be to get away from the solid core NTR and remove the requirement to keep the reactor solid, so you can use really high temperatures. A gas-core NTR could probably do SSTO quite easily.

[Robotbeat]

Nuclear thermal in general is worse than chemical for SSTO. Maybe an aggressive gas core would change that, but the efficiency would be lower than a chemical rocket. Kirk Sorensen has blogged it on selenianboondocks.com

Chemical rockets are more efficient than people think, and high performance nuclear thermal requires HEU which is much more expensive than natural gas per joule plus you need hydrogen propellant as well for high performance NTR, and you don’t even get the benefit of the 142MJ/kg embodied energy. Dumb.

Chemical rockets are literally better than NTR for RLVs.

[spacenut]

What about a fusion reactor engine that ejects plasma for thrust?  Only problem is the mass of electromagnets to contain the plasma in the combustion chamber and the nozzle.  I may have answered my own question, but I think it would use less hydrogen than a fission engine, thus maybe less mass for the spacecraft itself. 

Two stage to orbit seems to be the best solution right now, reusing the chemical first stage and developing a returnable second stage.  I do like Stoke Space's upper stage solution if it works well. 

[Robotbeat]

Note that fusion has too high of an Isp to be energy optimal.

People underestimate how good chemical rockets are. It takes 32MJ/kg to launch up into orbit assuming no losses of any kind, just kinetic and potential energy. A good chemical rocket like Falcon 9 already gets 320MJ/kg or so, and that’s including everything. If you had a truck carrying gravel up a hill, your chemical to potential energy conversion for that gravel is likely to be worse than 10%. Plus natural gas is like the cheapest energy source per joule that there is, other than pure sunlight and maybe LEU. Starship has the potential to get cost to orbit as low as transpacific air freight. It’s just not that much of a constraint.

If you want to get better, use rotating tethers. Then you can even beat the efficiency limit because you can reuse energy from reentry of another payload to launch assist a payload.

[Vultur]

Nuclear thermal in general is worse than chemical for SSTO. Maybe an aggressive gas core would change that, but the efficiency would be lower than a chemical rocket. Kirk Sorensen has blogged it on selenianboondocks.com

Chemical rockets are more efficient than people think, and high performance nuclear thermal requires HEU which is much more expensive than natural gas per joule plus you need hydrogen propellant as well for high performance NTR, and you don’t even get the benefit of the 142MJ/kg embodied energy. Dumb.

Chemical rockets are literally better than NTR for RLVs.

Absolutely, I don't think it makes any economic (or regulatory) sense, but in terms of the thread question of what it would take to make it possible - I think a gas core of some type is probably the best answer.

Even if you insist on SSTO, though, I think chemical would actually be easier.

[Starship Trooper]

What about a fusion reactor engine that ejects plasma for thrust?  Only problem is the mass of electromagnets to contain the plasma in the combustion chamber and the nozzle.

There are a raft of issues here, which the 1970s "Project Daedalus" design tried to address.  Note, this was a concept for a fusion powered interstellar probe.  The low thrust/mass ratio of such a fusion drive is not suitable for planetary ascents.  Wiki is good enough for an introductory synopsis: https://en.wikipedia.org/wiki/Project_Daedalus

Anyway, your magnets cannot contain the energetic neutrons produced by deuterium/tritium fusion.  So little propulsion would ensue and the neutrons would slowly disintegrate your spacecraft.  Therefore such concepts are based on deuterium/helium-3 fusion, which ejects energetic protons, which can be channeled by electromagnets.  Unfortunately, the fusion is exponentially harder to do:

Deuterium/Helium-3 (D-3He)

Difficulty of implementation. The D-3He reaction requires high temperatures and a limited temperature range compared to the D-T reaction. This is because the D-3He synthesis cross-section is lower than that of D-T, and the D-3He reaction requires more energy.

Problems with conditions. For example:
The need for high plasma density - The D-3He reaction requires a density product over a retention time that is 50 times higher than that of D-T.

No reliable source of helium-3 — helium-3 is a rare and expensive isotope that is not produced on an industrial scale. (Daedalus required 50,000 tonnes of fuel, of which ~ 30,000 tonnes would be helium-3.  The concept was to mine the atmosphere of a gas giant to get the 3He.  These guys thought big!)

Control complexity - Plasma retention and control of synthesis product accumulation are important for the D-3He reaction.

So basically, good luck with all of that.  The British boffins in the 1970s BIS thought implementation of such ideas might be half a century away.  Now they are looking to be over a century away, if ever ...

[sanman]

I'd once read that NASA has funded research into "Direct Fusion Drive" technology.

I recall we'd also had a thread on here about a Polywell fusion-based propulsion concept proposed by Bussard

Also, could something similar to Orion be done using laser-triggered Inertial Confinement Fusion of fuel pellets?

Note that fusion has too high of an Isp to be energy optimal.

But we should be able to use that high Isp to entrain additional propellant mass for increased thrust.
A ramjet design could have the high Isp fusion exhaust accelerating the incoming airflow.

[sanman]

What about a fusion reactor engine that ejects plasma for thrust?  Only problem is the mass of electromagnets to contain the plasma in the combustion chamber and the nozzle.

There are a raft of issues here, which the 1970s "Project Daedalus" design tried to address.  Note, this was a concept for a fusion powered interstellar probe.  The low thrust/mass ratio of such a fusion drive is not suitable for planetary ascents.  Wiki is good enough for an introductory synopsis: https://en.wikipedia.org/wiki/Project_Daedalus

Anyway, your magnets cannot contain the energetic neutrons produced by deuterium/tritium fusion.  So little propulsion would ensue and the neutrons would slowly disintegrate your spacecraft.  Therefore such concepts are based on deuterium/helium-3 fusion, which ejects energetic protons, which can be channeled by electromagnets.  Unfortunately, the fusion is exponentially harder to do:

Deuterium/Helium-3 (D-3He)

Difficulty of implementation. The D-3He reaction requires high temperatures and a limited temperature range compared to the D-T reaction. This is because the D-3He synthesis cross-section is lower than that of D-T, and the D-3He reaction requires more energy.

Problems with conditions. For example:
The need for high plasma density - The D-3He reaction requires a density product over a retention time that is 50 times higher than that of D-T.

No reliable source of helium-3 — helium-3 is a rare and expensive isotope that is not produced on an industrial scale. (Daedalus required 50,000 tonnes of fuel, of which ~ 30,000 tonnes would be helium-3.  The concept was to mine the atmosphere of a gas giant to get the 3He.  These guys thought big!)

Control complexity - Plasma retention and control of synthesis product accumulation are important for the D-3He reaction.

So basically, good luck with all of that.  The British boffins in the 1970s BIS thought implementation of such ideas might be half a century away.  Now they are looking to be over a century away, if ever ...

But how much 3He do you really need just for SSTO? We're not talking about supplying world energy needs. Let's suppose that if net energy-positive fusion were possible with 3He, then it gets prioritized for space applications, since it's supposed to be safer and less hazardous.

And what's so terrible about using Tritium, btw? It has a half-life of just 12 years, and you'd only need a small amount of it.

[Paul451]

Just to put it to bed, air-breathing launch is also "sounds good but doesn't make sense in practice". The trajectory for launch necessitates getting above the atmosphere as quickly as possible. You only gain a small amount of delta-v before you run out of air, during which you have to deal with high drag and thermal stress (plus the air-scoop is always heavy. Screwing your thrust/weight anyway.)

Nuclear plus air-breathing is piling garbage on garbage and wondering why it doesn't improve the smell.

I'd once read that NASA has funded research into "Direct Fusion Drive" technology.

Meh. NASA put money into the EM-drive.

Also, could something similar to Orion be done using laser-triggered Inertial Confinement Fusion of fuel pellets?

Have you seen the size of the laser-ICF reactors, versus the energy out? They make NTR reactors look dainty.

Note that fusion has too high of an Isp to be energy optimal.

But we should be able to use that high Isp to entrain additional propellant mass for increased thrust.

Sure. And after all that faffing around, you end up with something that has roughly the same thrust and Isp as a chemical rocket. Except the chemical rocket is 10% as heavy, costs 0.1% as much, is more reliable, and isn't spewing radiation.

Chemical fuels are optimal for Earth launch.

I know it's hard to get your head around, but it's true. We found the ideal case first. (The trick has been to get the construction and operating costs down to where the price of propellant is dominating the cost of launch.)

There are other applications where high-Isp, low-thrust rockets make sense. In which case, solar-electric is optimal out to the asteroid belt. After that, Jupiter and beyond, NEP becomes optimal. (Future tech, like fusion, might also compete with NEP, in that specific application.)

[Starship Trooper]

But how much 3He do you really need just for SSTO?

Mate, I love your interest in space, but I'm thinking your background is not STEM?  Anyway Daedalus was an optimistic minimum model of what a fusion reactor for spaceflight could be.  Dry mass of the reactor and craft, minus fuel and scientific payload, was 3500 tonnes.  The thrust was given as 7,540,000 newtons, which converts to ~ 769 tonnes of force.  So the thrust is less than a quarter of what would be needed to get the reactor off the ground, as I alluded to in my earlier post.  Fusion, if ever achieved, would be like ion drives for use in space only.  The low thrust is only redeemed by high ISP and the ability to thrust for years.

Tritium will not help you in any fashion.  You won't have thrust, just fast neutrons that are uncontrollable and so destructive they will limit the operational life of ground based D-T power reactors, if any are ever built.

[Starship Trooper]

Dear Paul,

You misinterpreted my off the cuff response to the OP's other question

"And what's so terrible about using Tritium, btw? It has a half-life of just 12 years, and you'd only need a small amount of it."

PS: Honors graduate of Caltech here, yes that was a very long time ago, but I'm not senile.  One of my proudest moments 'back in the day' was when "our" Voyagers were launched. And with 48 years of TLC, both are still working on reduced power and approaching a light-day out from Sol.  Back in the 1980s, there was a proposal for sending an ion drive probe out to the focus of the solar gravitational lens, over 3 light-days out.  Maybe people can start dreaming big again.

[sanman]

But how much 3He do you really need just for SSTO?

Mate, I love your interest in space, but I'm thinking your background is not STEM?  Anyway Daedalus was an optimistic minimum model of what a fusion reactor for spaceflight could be.  Dry mass of the reactor and craft, minus fuel and scientific payload, was 3500 tonnes.  The thrust was given as 7,540,000 newtons, which converts to ~ 769 tonnes of force.  So the thrust is less than a quarter of what would be needed to get the reactor off the ground, as I alluded to in my earlier post.  Fusion, if ever achieved, would be like ion drives for use in space only.  The low thrust is only redeemed by high ISP and the ability to thrust for years.

Yes, I understand that most of the devices used in mainstream fusion research are really big and heavy.
I wasn't talking about the mass of the machine, I was talking about how much of the rarified fuel you need to scrounge up for aneutronic purposes, since that's a concern with Helium-3.
Helium-3 is also produced as a by-product of nuclear reactors and weapons stockpiles, since it's a decay product. There's also the helium produced from the natural gas extraction industry, of which some small fraction is separated off by cryo-distillation as 3He.

Tritium will not help you in any fashion.  You won't have thrust, just fast neutrons that are uncontrollable and so destructive they will limit the operational life of ground based D-T power reactors, if any are ever built.

I understand that D-T fusion produces the extra neutron radiation, but that's a price you pay for the extra neutronic shielding effect to overcome Coulomb barrier.
Apologies, I wasn't trying to suggest using Tritium for D-T fusion, I was only mentioning Tritium because it too is a source of the rarified Helium-3, which is produced from Tritium decay. That's all I meant - sorry for the confusion.

Quick Google search says total worldwide annual production of 3He is 15-20kg.

So if nuclear fusion were to somehow be viable for space launch purposes, aren't our existing 3He sources enough to get us off the ground, without having to start strip-mining lunar surface or diving into Jupiter's upper atmosphere?

[Vultur]

Chemical fuels are optimal for Earth launch.

I know it's hard to get your head around, but it's true. We found the ideal case first. (The trick has been to get the construction and operating costs down to where the price of propellant is dominating the cost of launch.)

Yes, at least for internally powered vehicles. There's a possibility that some kind of high infrastructure approach like laser launch or Spinlaunch would make sense at a large enough volume of the right line of payloads, depending on how cheap & reliable the TPS maintenance on a reusable vehicle can be made.