3 ms·

As I understand it, the main waste product of concern is Uranium-232, which is highly radioactive (half-life of ~70 years) but not fissile. Conceivably you coul

As I understand it, the main waste product of concern is Uranium-232, which is highly radioactive (half-life of ~70 years) but not fissile. Conceivably you could maybe use it for power generation in some other way, though, like with a radioisotype thermoelectric generator? I'm not sure it would be worth the effort, though.

JoshTriplett 10y ago
I do find it sad that we don't yet have a more efficient means of capturing the output of a nuclear reactor than heating liquid to power a turbine. Seems like there should be a more direct (and more efficient) path from fission or fusion to electricity.
- lmm 10y ago
  Using the difference between a hot and cold reservoir to do work is a very fundamental, pure way to look at energy. It seems very natural to me.
  - rosser 10y ago
    I don't think the GP is bemoaning how "natural" the process seems, but how inefficient it is. It's pretty much the best we've got currently, given the choices we've made (including those broadly detailed in The Fine Article). That said, there's an aesthetic level at which you kinda have to look at the thermal efficiency of a nuclear power plant — which is, AFAIK, like 30-some percent — and wish we could somehow do better, given all the infrastructure and risk involved.
    - dv_dt 10y ago
      So has anyone ever calculated the effect of the heat output if all of our power consumption were supplied by nuclear power? (at 30% conversion efficiency). Would the earth just radiate that extra heat off, is it negligible for all practical purposes?
      - rosser 10y ago
        Probably not much different. I'm pretty sure most fossil fuel power plants have comparable efficiencies. It's not like the steam produced by burning dinosaurs is somehow less capable of driving a turbine than what comes from splitting atoms.
- mikeash 10y ago
  So much of the nuclear reaction ends up just wiggling the atoms around, so it's tough to do anything smarter than a heat engine. It is possible to generate power by other means. For example, you can use radioactive isotopes which emit beta radiation to generate electricity directly, since beta radiation is just free electrons. With fusion (ignoring the important problem of breakeven) you have an energetic plasma, and you can extract energy using magnetic fields rather than with turbines. For fission, the energy produced goes into moving neutrons and the fragments of the nuclei, which I don't think can really be captured other than as heat.
  - iaw 10y ago
    >> "ignoring the important problem of breakeven" I am so happy you included that line. From when I was studying fusion the actual conversion into usable energy was a complete after thought because breakeven is such a larger challenge.
    - mikeash 10y ago
      Indeed. It's an interesting question to figure out how to turn a sustainable fusion reaction into electricity, but it's totally pointless until the main problem is solved. In context I thought it was worth mentioning, since fusion does allow some interesting possibilities in this area, but there's no practical use for it at the moment.
  - hx87 10y ago
    MHD generators powered by decelerating nuclei fragments are possible and aren't subject to the limits of the Carnot Cycle, but they're still uneconomic at this point.
- CamperBob2 10y ago
  Most people underestimate just how insanely efficient steam really is. There aren't (m)any better ways of moving heat around and converting it into electricity, even after 2000 years.
  - sounds 10y ago
    I think he understood the efficiency limits, or in other words, the theory of the Carnot Cycle. Other than the minimal amount of global heating -- which should be thought about, but it's a lot better than CO2! -- why not just produce more reactors? We're not running out of Thorium, and maybe it can get us off coal. !!!
- neutronicus 10y ago
  Electricity moves energy by driving quasi-static fields with non-zero Poynting vector in a neighborhood of the wire. Heavy 2-MeV charged particles don't want to couple to that kind of mode at all; they're going to deposit their energy into MeV-ish radiation modes (e.g. Bremmstralung).
VLM 10y ago
U232 is not fun to work with, well into the "robot arms" level of gamma radiation output. U235 and P239 are like "wear gloves, glovebox" In a lab, or WRT inevitable accidents and contamination, its like the difference between working with strong industrial acids vs nerve gases. Its really a huge pain. The article author seems to have hand waved away several practical engineering problems like this. Also see his interesting hand waving away of the steam cycle as we know it, hand wave away molten salt moderator issues, etc. Some things are a practical pain because nobody's given the engineers enough $$$ yet, some things are a pain because of basic physics and chemistry reasons. I think the article author is confusing those two. Certainly, the nuclear industry over the last 70 years has not lacked for money or brainpower.

JoshTriplett 10y ago

I do find it sad that we don't yet have a more efficient means of capturing the output of a nuclear reactor than heating liquid to power a turbine. Seems like there should be a more direct (and more efficient) path from fission or fusion to electricity.

lmm 10y ago
Using the difference between a hot and cold reservoir to do work is a very fundamental, pure way to look at energy. It seems very natural to me.
- rosser 10y ago
  I don't think the GP is bemoaning how "natural" the process seems, but how inefficient it is. It's pretty much the best we've got currently, given the choices we've made (including those broadly detailed in The Fine Article). That said, there's an aesthetic level at which you kinda have to look at the thermal efficiency of a nuclear power plant — which is, AFAIK, like 30-some percent — and wish we could somehow do better, given all the infrastructure and risk involved.
  - dv_dt 10y ago
    So has anyone ever calculated the effect of the heat output if all of our power consumption were supplied by nuclear power? (at 30% conversion efficiency). Would the earth just radiate that extra heat off, is it negligible for all practical purposes?
    - rosser 10y ago
      Probably not much different. I'm pretty sure most fossil fuel power plants have comparable efficiencies. It's not like the steam produced by burning dinosaurs is somehow less capable of driving a turbine than what comes from splitting atoms.
mikeash 10y ago
So much of the nuclear reaction ends up just wiggling the atoms around, so it's tough to do anything smarter than a heat engine. It is possible to generate power by other means. For example, you can use radioactive isotopes which emit beta radiation to generate electricity directly, since beta radiation is just free electrons. With fusion (ignoring the important problem of breakeven) you have an energetic plasma, and you can extract energy using magnetic fields rather than with turbines. For fission, the energy produced goes into moving neutrons and the fragments of the nuclei, which I don't think can really be captured other than as heat.
- iaw 10y ago
  >> "ignoring the important problem of breakeven" I am so happy you included that line. From when I was studying fusion the actual conversion into usable energy was a complete after thought because breakeven is such a larger challenge.
  - mikeash 10y ago
    Indeed. It's an interesting question to figure out how to turn a sustainable fusion reaction into electricity, but it's totally pointless until the main problem is solved. In context I thought it was worth mentioning, since fusion does allow some interesting possibilities in this area, but there's no practical use for it at the moment.
- hx87 10y ago
  MHD generators powered by decelerating nuclei fragments are possible and aren't subject to the limits of the Carnot Cycle, but they're still uneconomic at this point.
CamperBob2 10y ago
Most people underestimate just how insanely efficient steam really is. There aren't (m)any better ways of moving heat around and converting it into electricity, even after 2000 years.
- sounds 10y ago
  I think he understood the efficiency limits, or in other words, the theory of the Carnot Cycle. Other than the minimal amount of global heating -- which should be thought about, but it's a lot better than CO2! -- why not just produce more reactors? We're not running out of Thorium, and maybe it can get us off coal. !!!
neutronicus 10y ago
Electricity moves energy by driving quasi-static fields with non-zero Poynting vector in a neighborhood of the wire. Heavy 2-MeV charged particles don't want to couple to that kind of mode at all; they're going to deposit their energy into MeV-ish radiation modes (e.g. Bremmstralung).

VLM 10y ago

U232 is not fun to work with, well into the "robot arms" level of gamma radiation output. U235 and P239 are like "wear gloves, glovebox" In a lab, or WRT inevitable accidents and contamination, its like the difference between working with strong industrial acids vs nerve gases. Its really a huge pain. The article author seems to have hand waved away several practical engineering problems like this. Also see his interesting hand waving away of the steam cycle as we know it, hand wave away molten salt moderator issues, etc. Some things are a practical pain because nobody's given the engineers enough $$$ yet, some things are a pain because of basic physics and chemistry reasons. I think the article author is confusing those two. Certainly, the nuclear industry over the last 70 years has not lacked for money or brainpower.