Why are NERVAs not yet used?

[Elthy]

Quote

Good. Now once we get people to stop protesting about nuclear power plants being built and get that Nevadan Nuclear Dump built, we can hope that NERVA will also be publically acceptable.  Remember, the public thinks the exhaust is radioactive (it's not). In that sense, Nuclear-Electric is likely better in the public view.

Those two are independent. Ive never heard anybody in Germany complain about nuclear power in space, and its propably the "anti-nuclear" country #1. Not choosing NERVA isnt about irrational fear of anything with "radiation", its a rational decission involing lots of factors, e.g. cost. R&D on such engines under current security standards wont be cheap.

[HebaruSan]

25 minutes ago, GreenWolf said:

@fredinno No, I'm not. ²³⁸Pu is fissile. The fact that you apparently don't understand what nuclear fission is doesn't make this any less true.

I'm the farthest thing from a nuclear physicist, but it does look like @fredinno was using the term correctly.

Quote

"Fissile" is distinct from "fissionable." A nuclide capable of undergoing fission (even with a low probability) after capturing a high energy neutron is referred to as "fissionable." A fissionable nuclide that can be induced to fission with low-energy thermal neutrons with a high probability is referred to as "fissile."^[3] Although the terms were formerly synonymous, fissionable materials include also those (such as uranium-238) that can be fissioned only with high-energy neutrons. As a result, fissile materials (such as uranium-235) are a subset of fissionable materials.

https://en.wikipedia.org/wiki/Fissile_material#Fissile_vs_fissionable

[SuperFastJellyfish]

23 minutes ago, GreenWolf said:

@fredinno No, I'm not. ²³⁸Pu is fissile. The fact that you apparently don't understand what nuclear fission is doesn't make this any less true.

I think you're mistaking fissile with fissionable.  https://en.wikipedia.org/wiki/Fissile_material#Fissile_vs_fissionable

From http://ieer.org/resource/factsheets/plutonium-factsheet/ :  

Quote

The even isotopes, plutonium-238, -240, and -242 are not fissile but yet are fissionable–that is, they can only be split by high energy neutrons. Generally, fissionable but non-fissile isotopes cannot sustain chain reactions; plutonium-240 is an exception to that rule.

 

[magnemoe]

22 minutes ago, Elthy said:

Those two are independent. Ive never heard anybody in Germany complain about nuclear power in space, and its propably the "anti-nuclear" country #1. Not choosing NERVA isnt about irrational fear of anything with "radiation", its a rational decission involing lots of factors, e.g. cost. R&D on such engines under current security standards wont be cheap.

This, most protesters are of the not in my backyard type, they don't care much about that happens far away.
Yes you have professional protesters but they are so few they can be ignored even if they get media time "for balance", they slide over into the nutty category like the lady who sued NASA because they impacted an asteroid. 
Anyway people has complained in newspapers about the beach having too much sand and its too many Spanish people around after being on holiday in Spain. 

[GreenWolf]

35 minutes ago, HebaruSan said:

I'm the farthest thing from a nuclear physicist...

Funny, I'm not. As in, I am an actual physicist. And as an actual physicist, I am telling you, I have never once, ever, heard anyone use the term fissionable. The definition that has been and is used is "an isotope capable of undergoing nuclear fission". The distinction between high-energy and low-energy neutrons is only useful in certain limited engineering circumstances, which I'm sure is where fissionable is used. But in the wider realm of general physics, the term is fissile.

Heck, the very same wikipedia article (the accuracy of which I find highly questionable) even says that the two terms are synonymous.

[Mad Rocket Scientist]

One other reason is that safely testing NERVAs is tricky and expensive, since you need to capture the exhaust.

[HebaruSan]

45 minutes ago, GreenWolf said:

Funny, I'm not. As in, I am an actual physicist. And as an actual physicist, I am telling you, I have never once, ever, heard anyone use the term fissionable. The definition that has been and is used is "an isotope capable of undergoing nuclear fission". The distinction between high-energy and low-energy neutrons is only useful in certain limited engineering circumstances, which I'm sure is where fissionable is used. But in the wider realm of general physics, the term is fissile.

Heck, the very same wikipedia article (the accuracy of which I find highly questionable) even says that the two terms are synonymous.

The Nuclear Regulatory Commission has an entry for it and seems to define the distinction the same way as the wiki article.

http://www.nrc.gov/reading-rm/basic-ref/glossary/fissionable-material.html

Quote

A nuclide that is capable of undergoing fission after capturing either high-energy (fast) neutrons or low-energy thermal (slow) neutrons. Although formerly used as a synonym for fissile material, fissionable materials also include those (such as uranium-238) that can be fissioned only with high-energy neutrons. As a result, fissile materials (such as uranium-235) are a subset of fissionable materials.

Uranium-235 fissions with low-energy thermal neutrons because the binding energy resulting from the absorption of a neutron is greater than the critical energy required for fission; therefore uranium-235 is a fissile material. By contrast, the binding energy released by uranium-238 absorbing a thermal neutron is less than the critical energy, so the neutron must possess additional energy for fission to be possible. Consequently, uranium-238 is a fissionable material.

 

[PB666]

7 minutes ago, HebaruSan said:

The Nuclear Regulatory Commission has an entry for it and seems to define the distinction the same way as the wiki article.

http://www.nrc.gov/reading-rm/basic-ref/glossary/fissionable-material.html

 

Lets not get this thread locked please. Remember there are defined and common uses of words.

[sevenperforce]

1 hour ago, Mad Rocket Scientist said:

One other reason is that safely testing NERVAs is tricky and expensive, since you need to capture the exhaust.

Errrr...why? Assuming that we're thinking about the same engine, there's no need to capture the exhaust. Irradiated hydrogen is not dangerous.

An NSWR, on the other hand, is just about impossible to test.

I'm still curious to know whether the uranium-core, lithium-hydride-propellant NTR would be feasible.

To the fissile/fissionable question: fissionable material is stuff like depleted uranium; fissile material is stuff like enriched uranium. Fissile material can form a critical mass; fissionable material cannot. They made bomb tampers out of non-fissile but still-fissionable material because that was a good way to increase yield without requiring more costly and dangerous fissile mass. Even though the non-fissile tamper could not go critical on its own, the neutron flux from the fissile core going critical would trigger fission in the tamper, dramatically boosting yield. This was most commonly used with thermonuclear weapons, and allowed the same basic primary and secondary to allow for a large range of variable yields based on tamper material.

All this is moot, of course, because I'm pretty sure there has never been a treaty about forbidding fissile or fissionable material in space; just a treaty banning nuclear weapons in space.

Edited February 9, 2016 by sevenperforce

[Mad Rocket Scientist]

27 minutes ago, sevenperforce said:

Errrr...why? Assuming that we're thinking about the same engine, there's no need to capture the exhaust. Irradiated hydrogen is not dangerous.

[...]

Here's where most of what I know about NERVAs came from: https://blogs.nasa.gov/J2X/2014/06/30/inside-the-leo-doghouse-nuclear-thermal-engines/

"Also, in addition to this, the hydrogen working fluid that we push through the reactor, it too picks up some level of radiation.  No, not a lot.  But under modern safety restrictions, all of that hydrogen would have to be captured and scrubbed clean before release."

[PB666]

On 2/8/2016 at 10:31 AM, DerekL1963 said:

There are three ways to reduce exposure:  Time, Distance, Shielding.

- Time is obviously not an option for a Mars journey, since it's years long.

- Distance is a big part of exposure reduction, as can be seen in pretty much any NTR craft design they put the crew as far away as humanly possible.  (Radiation follows the same inverse square law as visible light.)

- Shielding, well you don't need as much as you might think.   Not only does the vehicle's tankage and structure act as shielding, you only need a disk on the front of the reactor sufficient to put the rest of the vehicle into it's "shade". 

I don't see radiation of alpha, beta or gamma as a problem. A sheet of paper stops alpha, beta a little further, and the complex materials used to build a craft is not a real problem. Nuetron radiation can be a problem IMO during two phases. For example when in a equitorial orbit the craft once per orbit it traveling tail first in the direction of the sun, during this period neutrons will be emitted from the rocket nozzle and will continue to fly strait, however the solar wind can deflect them and send them back into the space craft. Boron infused metals in the craft hull can stop this, but during occupational EVA's they would not protect the nauts, nor protect them during transfers.

I agree with the poster in that there are concerns about long-term Use of Nerva. I think Nerva was proposed as a method for getting things way far away from earth, As in a New Horizons like launch were you do a major burn directly out of earths atmosphere. In this situation a craft reaches 95% of its orbital speed, decouples and then fires the nerva rocket placing it on an sol elliptical escape trajectory from earth. It has never been tested in a situation were the engine would be re-engaged 3 months later. My suspicion is that once the Nerva finishes its 3 hour burn, the engine and tank would be decoupled leaving a payload to travel on.

Liquid hydrogen . . . A typical 160 L liquid nitrogen tank boils off over 2 months, the larger the tank the smaller the surface area to mass ratio. Without convection the hydrogen boils more slowly. Since the tank is huge you can invest considerably more for insulation, HOWEVER, one has to be very careful because there is a tradeoff between strength and conductivity.

Let me give and example. lH₂ is extremely cold and expansion differentials between the tank and its extratank welds are stressed by the temperature differential. We can bolt the tank to a low conducting material such as carbon fiber frame for the tank, but the connects are going to be the same temperature as hydrogen unless the tank is of the double wall evacuated design (like a dewar), (which means that the most secured parts give little support to the metal with highest inertia, it is free to bounce with stresses. Space is liquids will not behave as in a dewar. If it is stored in a double walled actively evacuated container then periodically the hydrogen can be evacuated, recompressed or kept in recovery tanks were it can be used first. So thats not a big problem. The big problem is you need a massive double walled tank with little bottom support for a 2G launch that is also evacuated, creating even more pressure on the top of the tank. This could be dealt with if Carbon fiber standoffs are pressured into between the walls of the dewar. Again these will shrink so basically the tanks you have to be assembled weld by weld at very cold temperatures to kept cold until filled. So maybe somemore tech for lH2 needs to be accomplished to bring up that capacity.

Now we get to the issue of will hydrogen store until Mars portion of the missions is over. Lets think about this, if we could solve this problem we would not be talking about hyperglolic use for Mars lander, right. So its important. If we are talking about hydrogen as the primary ejection mass, its a problem, but if we are talking about hydrogen as a payload on SEP then during the transfer we have all the solar power in the world to reliquify hydrogen, so thats not a problem. If the Mars mission is a few day mission, your liquified gases should be OK (after all Mars irradiation is .4 of earth, its colder and its atmosphere is very thin). But here again if we are talking about a lfOx scenario on Mars (cause the Nerva does not have enough thrust to mass) then we already have a transfer engine and if we have a means of storing H we also have a means of storing O and so we don't need the added weight of a Nuclear engine. So now what is the best way to get from Earth LEO to Mars? for an ISP of 1000 maybe Nerva is not right, but what about ISP that can be scaled from 1250 to 9000.

My thoughts here are on the return we could use ION drives to ship even poorly stored fuels to Mars and then they could be liquefied and transferred in Mars orbit for the journey home.

Even at 3 or 4 times payload to shuttle weight the high efficiency SEP can transfer fuel as long as you allocate the time required to get from LEO to LMO (say 2 years). Again you could, theoretically use solar in the unused part of the flight to reliquify hydrogen indefinitely. You could literally park hydrogen in orbit or L1 of Mars and send gassy hydrogen on Ion driven shuttles for years waiting for a manned mission to support. The problem is that the refrigeration systems on the ISS have been a major source of repair issues. The parts are necessarily mechanical and wear. The bigger problem, unless you remove the CFC restrictions many of the modern cryogenic gases are not really designed with equipment longevity in mind (another phrase to describe them would not be appropriate). We had a -80 chest freezer from the 60s that we finally disposed of ~2000 it lasted about 35 years, both of the -80' Freezers we have bought failed just after the 5 year warranty was up and 8 to10 years is typical. For the really low temperature freezers you need to replace condenser line and the oil/gas separator every 5 years or the particulates from new-CFC degradation tend to clog up the lines. I know of people who take the old refrigerants and clean them up just to keep these old ultralows going, they are considerably more reliable than the new stuff. (seriously so few freezers are ultralows allowing the old-CFCs for this would not have even dented the Ozone layer, but . . . .). There are ways around the problem, for example more robust but wasteful seperators or back-up solenoid operated condensation lines, etc could create a fall back position.

 

 

[kunok]

30 minutes ago, sevenperforce said:

Errrr...why? Assuming that we're thinking about the same engine, there's no need to capture the exhaust. Irradiated hydrogen is not dangerous.

The insulation of nuclear fuel it's not perfect so the exhaust has parts of it.

[Bill Phil]

1 hour ago, kunok said:

The insulation of nuclear fuel it's not perfect so the exhaust has parts of it.

Even so, it's not much at all.

Much more dangerous is the radiation emitter in all directions. This can be solved with a shadow shield, but that doesn't completeky solve the problem.

[_Augustus_]

It's because unfortunately most of the current people in charge of our government grew up associating "nuclear" with "weapon of mass annihilation" or "that nasty reactor that blew up and mutated people".

Edited February 10, 2016 by _Augustus_

[Stargate525]

Because for some inexplicable reason, we don't trust other countries launching large amounts of nuclear material into space with a 'trust us it's not a weapon.' Can't possibly imagine why they would be skittish about that...

[mikegarrison]

OK, after three pages of "we're so awesome because we aren't scared of nuclear power" chest-thumping, IMO none of that has anything to do with the original question. The answer is what was alluded to by a few other posters: what would we use them FOR? Can you name a mission that has been actually flown for which a nuclear-thermal rocket would have been the best design? (Best including cheapest, most reliable, etc.)

[DBowman]

1 hour ago, mikegarrison said:

what would we use them FOR? Can you name a mission that has been actually flown for which a nuclear-thermal rocket would have been the best design? (Best including cheapest, most reliable, etc.)

I don't know that we've 'actually flown' a mission for which they 'd be the best design. NASA DRA ( man on Mars, #5 is the one I've read, the pdf is linked at the bottom of that page) evaluated (chemical, NTR) * (propulsive, aerocapture) I think NTR 'won' as far as minimizing mass required to LEO.

I guess we'd use them for man on Mars (or any other high mass high deltaV minimize time mission profile) if mass to LEO were the only consideration - but we're not going to use them if we are not flying those kind of profiles.

Edited February 10, 2016 by DBowman

[magnemoe]

3 hours ago, mikegarrison said:

OK, after three pages of "we're so awesome because we aren't scared of nuclear power" chest-thumping, IMO none of that has anything to do with the original question. The answer is what was alluded to by a few other posters: what would we use them FOR? Can you name a mission that has been actually flown for which a nuclear-thermal rocket would have been the best design? (Best including cheapest, most reliable, etc.)

Good question, it was an option for the Pluto flyby then they did the plausibility study to see if it could be done.
It could have helped during the Apollo moon missions, however here radiation would be an serious issue because of the small ship. An larger reusable transfer tug was not an option during Apollo as it would introduce far more complications. 

Still an reusable moon tug probably be an perfect place for nerva now that its looks its an increased focus on manned moon missions.
It has some benefits, the obvious better isp, still nerva has benefit over other advanced engines here, its has decent trust so it will not increase travel time much and not too long in the van allen belt. 
Need to store hydrogen but not very long, plenty of abort options if something goes wrong. 

[sevenperforce]

What about using something like ethane as the working fluid? Liquid at room temperature, plenty of hydrogen, not too much carbon.

 

[Frozen_Heart]

18 hours ago, Mad Rocket Scientist said:

One other reason is that safely testing NERVAs is tricky and expensive, since you need to capture the exhaust.

The exhaust isn't radioactive in most fission rocket designs. Especially with the solid core rockets that were being tested in the nerva program. Its just superheated hydrogen.

If the fuel rods are poorly made or in certain liquid and gas core designs the exhaust can be radiative, but they have never really been pursued.

[tater]

4 minutes ago, Frozen_Heart said:

The exhaust isn't radioactive in most fission rocket designs. Especially with the solid core rockets that were being tested in the nerva program. Its just superheated hydrogen.

If the fuel rods are poorly made or in certain liquid and gas core designs the exhaust can be radiative, but they have never really been pursued.

This is strictly speaking not true, particularly in the original NERVA design. They lost about 17kg of the solid core due to ablation as I recall. The later work in nearby (to me  ) Los Alamos solved most of those problems, however, and newer designs indeed have very little radioactive material lost in the exhaust. You are right of course about the hydrogen being just fine.

Edited February 10, 2016 by tater

[fredinno]

18 hours ago, PB666 said:

I don't see radiation of alpha, beta or gamma as a problem. A sheet of paper stops alpha, beta a little further, and the complex materials used to build a craft is not a real problem. Nuetron radiation can be a problem IMO during two phases. For example when in a equitorial orbit the craft once per orbit it traveling tail first in the direction of the sun, during this period neutrons will be emitted from the rocket nozzle and will continue to fly strait, however the solar wind can deflect them and send them back into the space craft. Boron infused metals in the craft hull can stop this, but during occupational EVA's they would not protect the nauts, nor protect them during transfers.

I agree with the poster in that there are concerns about long-term Use of Nerva. I think Nerva was proposed as a method for getting things way far away from earth, As in a New Horizons like launch were you do a major burn directly out of earths atmosphere. In this situation a craft reaches 95% of its orbital speed, decouples and then fires the nerva rocket placing it on an sol elliptical escape trajectory from earth. It has never been tested in a situation were the engine would be re-engaged 3 months later. My suspicion is that once the Nerva finishes its 3 hour burn, the engine and tank would be decoupled leaving a payload to travel on.

Liquid hydrogen . . . A typical 160 L liquid nitrogen tank boils off over 2 months, the larger the tank the smaller the surface area to mass ratio. Without convection the hydrogen boils more slowly. Since the tank is huge you can invest considerably more for insulation, HOWEVER, one has to be very careful because there is a tradeoff between strength and conductivity.

Let me give and example. lH₂ is extremely cold and expansion differentials between the tank and its extratank welds are stressed by the temperature differential. We can bolt the tank to a low conducting material such as carbon fiber frame for the tank, but the connects are going to be the same temperature as hydrogen unless the tank is of the double wall evacuated design (like a dewar), (which means that the most secured parts give little support to the metal with highest inertia, it is free to bounce with stresses. Space is liquids will not behave as in a dewar. If it is stored in a double walled actively evacuated container then periodically the hydrogen can be evacuated, recompressed or kept in recovery tanks were it can be used first. So thats not a big problem. The big problem is you need a massive double walled tank with little bottom support for a 2G launch that is also evacuated, creating even more pressure on the top of the tank. This could be dealt with if Carbon fiber standoffs are pressured into between the walls of the dewar. Again these will shrink so basically the tanks you have to be assembled weld by weld at very cold temperatures to kept cold until filled. So maybe somemore tech for lH2 needs to be accomplished to bring up that capacity.

Now we get to the issue of will hydrogen store until Mars portion of the missions is over. Lets think about this, if we could solve this problem we would not be talking about hyperglolic use for Mars lander, right. So its important. If we are talking about hydrogen as the primary ejection mass, its a problem, but if we are talking about hydrogen as a payload on SEP then during the transfer we have all the solar power in the world to reliquify hydrogen, so thats not a problem. If the Mars mission is a few day mission, your liquified gases should be OK (after all Mars irradiation is .4 of earth, its colder and its atmosphere is very thin). But here again if we are talking about a lfOx scenario on Mars (cause the Nerva does not have enough thrust to mass) then we already have a transfer engine and if we have a means of storing H we also have a means of storing O and so we don't need the added weight of a Nuclear engine. So now what is the best way to get from Earth LEO to Mars? for an ISP of 1000 maybe Nerva is not right, but what about ISP that can be scaled from 1250 to 9000.

My thoughts here are on the return we could use ION drives to ship even poorly stored fuels to Mars and then they could be liquefied and transferred in Mars orbit for the journey home.

Even at 3 or 4 times payload to shuttle weight the high efficiency SEP can transfer fuel as long as you allocate the time required to get from LEO to LMO (say 2 years). Again you could, theoretically use solar in the unused part of the flight to reliquify hydrogen indefinitely. You could literally park hydrogen in orbit or L1 of Mars and send gassy hydrogen on Ion driven shuttles for years waiting for a manned mission to support. The problem is that the refrigeration systems on the ISS have been a major source of repair issues. The parts are necessarily mechanical and wear. The bigger problem, unless you remove the CFC restrictions many of the modern cryogenic gases are not really designed with equipment longevity in mind (another phrase to describe them would not be appropriate). We had a -80 chest freezer from the 60s that we finally disposed of ~2000 it lasted about 35 years, both of the -80' Freezers we have bought failed just after the 5 year warranty was up and 8 to10 years is typical. For the really low temperature freezers you need to replace condenser line and the oil/gas separator every 5 years or the particulates from new-CFC degradation tend to clog up the lines. I know of people who take the old refrigerants and clean them up just to keep these old ultralows going, they are considerably more reliable than the new stuff. (seriously so few freezers are ultralows allowing the old-CFCs for this would not have even dented the Ozone layer, but . . . .). There are ways around the problem, for example more robust but wasteful seperators or back-up solenoid operated condensation lines, etc could create a fall back position.

 

 

You can;t use conventional solid cores and get 9000 Isp- you need something like Liquid core, which releases lots of radiation out the other end.

 

11 hours ago, magnemoe said:

Good question, it was an option for the Pluto flyby then they did the plausibility study to see if it could be done.
It could have helped during the Apollo moon missions, however here radiation would be an serious issue because of the small ship. An larger reusable transfer tug was not an option during Apollo as it would introduce far more complications. 

Still an reusable moon tug probably be an perfect place for nerva now that its looks its an increased focus on manned moon missions.
It has some benefits, the obvious better isp, still nerva has benefit over other advanced engines here, its has decent trust so it will not increase travel time much and not too long in the van allen belt. 
Need to store hydrogen but not very long, plenty of abort options if something goes wrong. 

Why would you develop NERVA just to flyby Pluto? And the only time you'd want to use NERVA is for crew to escape the Van Allen quickly. Only problem is, that's a very small amount of stuff. ION drives last a lot longer, don't have disposal problems, and can be used for GEO missions too, allowing something like that to be commercially viable.

For most of the mass

3 hours ago, sevenperforce said:

What about using something like ethane as the working fluid? Liquid at room temperature, plenty of hydrogen, not too much carbon.

 

I'm pretty sure just some carbon still is pretty bad for a NERVA, reducing its lifetime significantly.

[sevenperforce]

1 hour ago, fredinno said:

I'm pretty sure just some carbon still is pretty bad for a NERVA, reducing its lifetime significantly.

Ah, I see. Well, in that case I like the self-pressurizing lithium-hydride-saltwater approach, with fuel fission.

Or we could go for something more exotic...like a gas-core centrifugal self-pumping neutron laser. Talk about a torchship....

EDIT:

Here's what I mean.

The entire thing is made of a neutron-reflective material, except for the hole at one end, which is plugged with a neutron-transparent material. You pump it full of uranium in whatever gaseous state you can manage. The casing has a magnetic charge, so you can spin it up by simply applying an external field.

Once it starts spinning fast enough (clockwise, mind you), the gaseous uranium reaches sufficient density in the deep outer channels to go critical. This produces heat and pressure in those channels, pushing them in the direction of rotation and accelerating the reaction.

High-energy neutrons produced by the reaction are aligned by the internal magnetic field of the rotating centrifuge and either bounce off the reflector at one end or pass through the neutron-transparent plug at the other end to escape.

Effective exhaust velocity: 20,000 km/s. 

Edited February 10, 2016 by sevenperforce
adding info

[PB666]

3 hours ago, sevenperforce said:

Ah, I see. Well, in that case I like the self-pressurizing lithium-hydride-saltwater approach, with fuel fission.

Or we could go for something more exotic...like a gas-core centrifugal self-pumping neutron laser. Talk about a torchship....

EDIT:

Here's what I mean.

The entire thing is made of a neutron-reflective material, except for the hole at one end, which is plugged with a neutron-transparent material. You pump it full of uranium in whatever gaseous state you can manage. The casing has a magnetic charge, so you can spin it up by simply applying an external field.

Once it starts spinning fast enough (clockwise, mind you), the gaseous uranium reaches sufficient density in the deep outer channels to go critical. This produces heat and pressure in those channels, pushing them in the direction of rotation and accelerating the reaction.

High-energy neutrons produced by the reaction are aligned by the internal magnetic field of the rotating centrifuge and either bounce off the reflector at one end or pass through the neutron-transparent plug at the other end to escape.

Effective exhaust velocity: 20,000 km/s. 

Why would neutrons align in a magnetic field, and second why would they reflect? I know neutrons are deflected by certain nuclei, I am not aware that anything reflects them like a mirror.

[sevenperforce]

15 hours ago, PB666 said:

Why would neutrons align in a magnetic field, and second why would they reflect? I know neutrons are deflected by certain nuclei, I am not aware that anything reflects them like a mirror.

Neutrons have a magnetic moment, so a rotating magnetic field can almost certainly be configured to align them. Deflection by the appropriate nuclei is close enough to reflection to work, I think.