4 ms·

Nuclear has the problem that you somehow need to hoist a reactor from the Earth into space, and putting nuclear materials makes a lot of people nervous. (What

by elihu 4y ago

Nuclear has the problem that you somehow need to hoist a reactor from the Earth into space, and putting nuclear materials makes a lot of people nervous. (What happens if the rocket launch fails and scatters radioactive debris?) An alternative is to gather the reactor fuel on Mars, but that sounds like a difficult undertaking. I think the expectation is that the energy needed to make methane on Mars will be provided by a lot of solar panels. (I would be curious to know how the math works out. For instance, if you send a Starship fully loaded with solar panels to Mars and spread them out on the ground, how long will it take those solar panels to gather enough energy to make enough methane to return the rocket to Earth? Is it one year? 10 years? 100 years?)

trhway 4y ago
The amount of highly enriched U and Pu that has already been put into space is at least several hundred of kilograms across at least 10+ of launches. And reliability of SpaceX launches is highest in history of human space programs. My bet is on nuclear vs solar as the nuclear is the next rocket engine type after chemical and will loose the dependency on launch windows and shorten the trip time.
- elihu 4y ago
  The safety risk isn't necessarily an insurmountable obstacle; I could imagine the U.S. government being more likely to approve sending a nuclear reactor into space if it was NASA that was asking as opposed to a private company, but either way it's more of a political problem than a technical one. Nuclear rocket engines are tricky. Nuclear reactors don't actually work very well in space because there isn't any convenient way to get rid of the excess heat. A vacuum is a very good insulator, so usually your only option is just to radiate it away as infrared light. With a rocket there's another option which is to transfer all the heat to the reaction mass you're expelling out the back of the ship. That sounds like a hard engineering problem though. Using a reactor on the Mars surface is a lot more straightforward because you can use the local air and ground to transfer heat. And since Mars is so cold, you might even get better steam generator efficiency there than on Earth, where the ambient temperatures are higher. One of the hopes with fusion is that if it pans out it might be reasonable to send a fusion reactor to Mars since you wouldn't need to send radioactive fuel rods. In fact, maybe the first practical fusion reactors will be used on Mars before they're used in more than a demonstrative capacity on Earth because they fit a very specific need, there are barriers to using the alternatives, and cost per kw/h isn't the most important constraint.
  - XorNot 4y ago
    Project Pluto: https://en.wikipedia.org/wiki/Project_Pluto https://en.wikipedia.org/wiki/Project_Pluto The US did build a nuclear powered ramjet engine and test it. This isn't an insurmountable hurdle, it's an achieved one. The design they used of course has lots of good reasons we should not use it on Earth (namely, the reactor would activate the air, meaning it spews a plume of oxygen and nitrogen isotopes) but it does work. You can solve the radionucleotide problem by running it off hydrogen instead, which doesn't activate in any meaningful quantity. Basically, not only do we have the ability, we had it in 1961.
    - elihu 4y ago
      Interesting. That requires an atmosphere though; it might work for the first stage of a launch from Earth, and maybe even for Mars. Once you're out of the atmosphere you'd need some other method of propulsion.
      - trhway 4y ago
        I think nuclear space propulsion would be used as a reactor generating energy for ionic drive. That would work as a Mars-Earth shuttle while never going into atmosphere. The landing to and launching from the plane surface would be done by regular chemical rockets.
      - XorNot 4y ago
        Specifically it requires propellant, which on Earth is the air intake. For space travel you have to still take a propellant, but it can be essentially any type of gaseous reaction mass - for example, water.
alex_young 4y ago
There are plenty of nuclear reactors already in space. The Voyager probes are famously powered by radioisotope thermoelectric generators, and even the Perseverance lander mentioned in the article is powered by one. Full list here: https://en.wikipedia.org/wiki/List_of_nuclear_power_systems_in_space https://en.wikipedia.org/wiki/List_of_nuclear_power_systems_...
- aqme28 4y ago
  Those are much much smaller than the nuclear reactors that parent comment is talking about. A totally different beast.
  - elihu 4y ago
    Interestingly the biggest reactors (mostly 2kw and a few 5kw) were deployed by the Soviet Union. Presumably they just didn't care about the safety risks. The risks weren't purely theoretical either. They had one nuclear-powered satellite that re-entered over Canada. https://en.wikipedia.org/wiki/Kosmos_954 https://en.wikipedia.org/wiki/Kosmos_954
- midoridensha 4y ago
  Those aren't "reactors". They're just lumps of plutonium that are decaying naturally and producing heat. A reactor has to be able to control a fission reaction; these can't. https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_generator https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_ge...
PaulHoule 4y ago
On mars you have a day not too different from ours and solar looks attractive. On the moon you really need nuclear because the night lasts two weeks. It is so bad people are thinking about reversing O’Neills idea and microwave beaming power from the Earth to the Moon.
- midoridensha 4y ago
  Why would you need to do that? Just put solar arrays at the poles. There might be some practical limitations with power transmission to other parts of the moon, but extremely high-voltage power should be easier to do there than on Earth, since there's no oceans in the way, nor any atmosphere (so no problems with discharge from the high voltage wires).
rdevsrex 4y ago
Couldn't we just use low enriched uranium for that? And as long as you aren't actively powering it up on takeoff, even in an accident, how radioactive would it be?
- elihu 4y ago
  That's a good question, I don't know enough to answer. I suppose one way to mitigate the risk is to use the same kind of "emergency escape" systems that humans use during rocket launches. I.e. if the rocket fails to achieve orbit, a capsule is ejected and floats down on parachutes. I wonder if that sort of thing is already done with RTG launches?
  - ncmncm 4y ago
    It is not. But they put it in a sturdy can.
dmurray 4y ago
> I would be curious to know how the math works out...how long will it take those solar panels to gather enough energy to make enough methane to return the rocket to Earth? Somewhere around a year. A Starship, refuelled in Earth orbit, intends to carry 100 tons to Mars [0] using 1200 tons of propellant [1] with an oxidizer/fuel ratio of 78:22 [1], so 265 tons of methane. Spacex's numbers are likely measured from low Earth orbit to low Mars orbit and assuming aerobraking on the way down. According to [2] that takes 5.8 km/s delta-v, but it's an additional 3.8 km/s to travel from the Mars surface. That's 65% extra, meaning we will need e^0.65 as much fuel because of [3], which is 1.9x, so 500 tons of methane. So the first answer is, it can't be done by refueling a Starship final stage alone - you'd need a booster, or (simpler) reduce payload by half. I'll assume instead we divided the load between two rockets coming back. Methane has a specific energy of 55MJ/kg, [4] so we need 28 TJ of energy to fill the tank. That's 8 GWh. Mars receives about 50% as much sunlight as Earth [5] Spacefaring solar panels weigh 2kg/sqm and produce 350W/sqm in Earth orbit. [6]. Terrestrial solar panels are cheaper to produce, but heavier. That's baseplate capacity, so make it 1/3 of that to average over day and night, and halve it again for running at Mars light levels, giving 60W/sqm or 30W/kg. 100 tons of that is 3 MW. Typical conversion efficiency for power-to-methane is 50%. [7]. This is done via hydrogen and requires carbon dioxide, which is abundant in the Mars atmosphere, and water, which is not, but is hopefully available in the quantities we require. So our solar plant produces 1.5MW of methane. I'm going to assume we get our oxygen for free as a by-product without any loss of efficiency, which seems plausible to me but not well documented. 8 GWh/1.5 MW is 5,300 hours or 225 days (either type) or about 8 earth months. I haven't allowed for the mass cost of the infrastructure for methane refining, or degradation of solar panels in the Mars dust. So all this should be a considered a lower bound for your answer. [0] https://www.spacex.com/human-spaceflight/mars/ https://www.spacex.com/human-spaceflight/mars/ [1] https://www.wevolver.com/specs/spacexs-starship-sn24-bn7 https://www.wevolver.com/specs/spacexs-starship-sn24-bn7 [2] https://i.imgur.com/SqdzxzF.png https://i.imgur.com/SqdzxzF.png [3] https://en.wikipedia.org/wiki/Tsiolkovsky_rocket_equation https://en.wikipedia.org/wiki/Tsiolkovsky_rocket_equation [4] https://en.wikipedia.org/wiki/Energy_density https://en.wikipedia.org/wiki/Energy_density [5] https://www.quora.com/How-bright-is-sunlight-on-Mars https://www.quora.com/How-bright-is-sunlight-on-Mars [6] https://www.spectrolab.com/DataSheets/Panel/panels.pdf https://www.spectrolab.com/DataSheets/Panel/panels.pdf [7] https://www.sciencedirect.com/science/article/pii/S0306261918311826 https://www.sciencedirect.com/science/article/pii/S030626191...

by elihu 4y ago

trhway 4y ago

The amount of highly enriched U and Pu that has already been put into space is at least several hundred of kilograms across at least 10+ of launches. And reliability of SpaceX launches is highest in history of human space programs. My bet is on nuclear vs solar as the nuclear is the next rocket engine type after chemical and will loose the dependency on launch windows and shorten the trip time.

elihu 4y ago
The safety risk isn't necessarily an insurmountable obstacle; I could imagine the U.S. government being more likely to approve sending a nuclear reactor into space if it was NASA that was asking as opposed to a private company, but either way it's more of a political problem than a technical one. Nuclear rocket engines are tricky. Nuclear reactors don't actually work very well in space because there isn't any convenient way to get rid of the excess heat. A vacuum is a very good insulator, so usually your only option is just to radiate it away as infrared light. With a rocket there's another option which is to transfer all the heat to the reaction mass you're expelling out the back of the ship. That sounds like a hard engineering problem though. Using a reactor on the Mars surface is a lot more straightforward because you can use the local air and ground to transfer heat. And since Mars is so cold, you might even get better steam generator efficiency there than on Earth, where the ambient temperatures are higher. One of the hopes with fusion is that if it pans out it might be reasonable to send a fusion reactor to Mars since you wouldn't need to send radioactive fuel rods. In fact, maybe the first practical fusion reactors will be used on Mars before they're used in more than a demonstrative capacity on Earth because they fit a very specific need, there are barriers to using the alternatives, and cost per kw/h isn't the most important constraint.
- XorNot 4y ago
  Project Pluto: https://en.wikipedia.org/wiki/Project_Pluto https://en.wikipedia.org/wiki/Project_Pluto The US did build a nuclear powered ramjet engine and test it. This isn't an insurmountable hurdle, it's an achieved one. The design they used of course has lots of good reasons we should not use it on Earth (namely, the reactor would activate the air, meaning it spews a plume of oxygen and nitrogen isotopes) but it does work. You can solve the radionucleotide problem by running it off hydrogen instead, which doesn't activate in any meaningful quantity. Basically, not only do we have the ability, we had it in 1961.
  - elihu 4y ago
    Interesting. That requires an atmosphere though; it might work for the first stage of a launch from Earth, and maybe even for Mars. Once you're out of the atmosphere you'd need some other method of propulsion.
    - trhway 4y ago
      I think nuclear space propulsion would be used as a reactor generating energy for ionic drive. That would work as a Mars-Earth shuttle while never going into atmosphere. The landing to and launching from the plane surface would be done by regular chemical rockets.
    - XorNot 4y ago
      Specifically it requires propellant, which on Earth is the air intake. For space travel you have to still take a propellant, but it can be essentially any type of gaseous reaction mass - for example, water.

alex_young 4y ago

There are plenty of nuclear reactors already in space. The Voyager probes are famously powered by radioisotope thermoelectric generators, and even the Perseverance lander mentioned in the article is powered by one. Full list here: https://en.wikipedia.org/wiki/List_of_nuclear_power_systems_in_space https://en.wikipedia.org/wiki/List_of_nuclear_power_systems_...

aqme28 4y ago
Those are much much smaller than the nuclear reactors that parent comment is talking about. A totally different beast.
- elihu 4y ago
  Interestingly the biggest reactors (mostly 2kw and a few 5kw) were deployed by the Soviet Union. Presumably they just didn't care about the safety risks. The risks weren't purely theoretical either. They had one nuclear-powered satellite that re-entered over Canada. https://en.wikipedia.org/wiki/Kosmos_954 https://en.wikipedia.org/wiki/Kosmos_954
midoridensha 4y ago
Those aren't "reactors". They're just lumps of plutonium that are decaying naturally and producing heat. A reactor has to be able to control a fission reaction; these can't. https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_generator https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_ge...

PaulHoule 4y ago

On mars you have a day not too different from ours and solar looks attractive. On the moon you really need nuclear because the night lasts two weeks. It is so bad people are thinking about reversing O’Neills idea and microwave beaming power from the Earth to the Moon.

midoridensha 4y ago
Why would you need to do that? Just put solar arrays at the poles. There might be some practical limitations with power transmission to other parts of the moon, but extremely high-voltage power should be easier to do there than on Earth, since there's no oceans in the way, nor any atmosphere (so no problems with discharge from the high voltage wires).

rdevsrex 4y ago

Couldn't we just use low enriched uranium for that? And as long as you aren't actively powering it up on takeoff, even in an accident, how radioactive would it be?

elihu 4y ago
That's a good question, I don't know enough to answer. I suppose one way to mitigate the risk is to use the same kind of "emergency escape" systems that humans use during rocket launches. I.e. if the rocket fails to achieve orbit, a capsule is ejected and floats down on parachutes. I wonder if that sort of thing is already done with RTG launches?
- ncmncm 4y ago
  It is not. But they put it in a sturdy can.

dmurray 4y ago

> I would be curious to know how the math works out...how long will it take those solar panels to gather enough energy to make enough methane to return the rocket to Earth? Somewhere around a year. A Starship, refuelled in Earth orbit, intends to carry 100 tons to Mars [0] using 1200 tons of propellant [1] with an oxidizer/fuel ratio of 78:22 [1], so 265 tons of methane. Spacex's numbers are likely measured from low Earth orbit to low Mars orbit and assuming aerobraking on the way down. According to [2] that takes 5.8 km/s delta-v, but it's an additional 3.8 km/s to travel from the Mars surface. That's 65% extra, meaning we will need e^0.65 as much fuel because of [3], which is 1.9x, so 500 tons of methane. So the first answer is, it can't be done by refueling a Starship final stage alone - you'd need a booster, or (simpler) reduce payload by half. I'll assume instead we divided the load between two rockets coming back. Methane has a specific energy of 55MJ/kg, [4] so we need 28 TJ of energy to fill the tank. That's 8 GWh. Mars receives about 50% as much sunlight as Earth [5] Spacefaring solar panels weigh 2kg/sqm and produce 350W/sqm in Earth orbit. [6]. Terrestrial solar panels are cheaper to produce, but heavier. That's baseplate capacity, so make it 1/3 of that to average over day and night, and halve it again for running at Mars light levels, giving 60W/sqm or 30W/kg. 100 tons of that is 3 MW. Typical conversion efficiency for power-to-methane is 50%. [7]. This is done via hydrogen and requires carbon dioxide, which is abundant in the Mars atmosphere, and water, which is not, but is hopefully available in the quantities we require. So our solar plant produces 1.5MW of methane. I'm going to assume we get our oxygen for free as a by-product without any loss of efficiency, which seems plausible to me but not well documented. 8 GWh/1.5 MW is 5,300 hours or 225 days (either type) or about 8 earth months. I haven't allowed for the mass cost of the infrastructure for methane refining, or degradation of solar panels in the Mars dust. So all this should be a considered a lower bound for your answer. [0] https://www.spacex.com/human-spaceflight/mars/ https://www.spacex.com/human-spaceflight/mars/ [1] https://www.wevolver.com/specs/spacexs-starship-sn24-bn7 https://www.wevolver.com/specs/spacexs-starship-sn24-bn7 [2] https://i.imgur.com/SqdzxzF.png https://i.imgur.com/SqdzxzF.png [3] https://en.wikipedia.org/wiki/Tsiolkovsky_rocket_equation https://en.wikipedia.org/wiki/Tsiolkovsky_rocket_equation [4] https://en.wikipedia.org/wiki/Energy_density https://en.wikipedia.org/wiki/Energy_density [5] https://www.quora.com/How-bright-is-sunlight-on-Mars https://www.quora.com/How-bright-is-sunlight-on-Mars [6] https://www.spectrolab.com/DataSheets/Panel/panels.pdf https://www.spectrolab.com/DataSheets/Panel/panels.pdf [7] https://www.sciencedirect.com/science/article/pii/S0306261918311826 https://www.sciencedirect.com/science/article/pii/S030626191...