Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Nanomaterial Converts Radiation to Electricity

  1. Mar 31, 2008 #1
    A nanomaterial has been designed to convert radiation directly into electricity:


    It is said to be upto 20 times more efficient than radioisotope thermoelectric generation.


    This sounds like 1 MW / L !! :bugeye:

    I'm wondering if something like this could be used to power a VASIMR rocket?

    What else could it power? Dune-buggy sized Mars rovers? Space probes are obvious, though.

    What about things down here on Earth?
    Antarctic land transport? Nuclear-powered pacemakers?
    Mountaintop communication relays?

    Battlefield lasers?
    Plasma-assisted drag reduction for high-speed aerospace craft?
    Last edited: Mar 31, 2008
  2. jcsd
  3. Mar 31, 2008 #2
    What should we call this technology exactly?

  4. Mar 31, 2008 #3


    User Avatar
    Science Advisor

    If it were scaled up, could this here wang-doodle be more efficient than a nuclear power-plant at converting radioactive fuel to electricity?
  5. Mar 31, 2008 #4


    User Avatar
    Gold Member

    This is promising as a new energy conversion technology; its not an energy source. For that you'll still need the nuclear radiation producing source, and the only thing that produces that at significant power levels is still controlled nuclear fission.
  6. Mar 31, 2008 #5


    User Avatar
    Gold Member

    That seems to the vagues suggestion in the links - that this would be a big improvement over a heat cycle thats limited by thermodynamics. Its hard to tell though as there aren't many details. Twenty times better than what? Article doesn't say. What kind of radiation is required? Im guessing alphas, maybe even gamma, anything but neutrons.
  7. Mar 31, 2008 #6
    Well, consider that with a regular nuclear power plant, you've got a lot more conversion steps to get to electricity -- and that means more losses along the way.

    You've got to transfer the heat from the nuclear pile to the coolant or transfer medium, then that's going to transfer to another medium like water, so that you can get steam to turn a turbine, which turns a generator to get the electricity.

    This new nanomaterial directly turns radiation into electricity, and sucks it out via the nanotubes. They're saying it could be 20 times more efficient than the old RTGs, which are nice, simple and reliable in design, but have low efficiency. (3-7%)


    The nanotubes are said to be robust, because even if the ionizing radiation damages them, electromigration from the current they're carrying will tend to repair the nanotubes and re-form them back. So the material is less likely to degrade over time.

    Here, take a look at this:


    Silicon Carbide was also considered promising for the N-battery thing. But the radiation quickly causes the material to degrade. The current-carrying nanotubes are more resilient to this.

    Going back to the launch vehicle thing again, what about a particle-bed concept for the N-battery? Particle-bed N-reactors are supposed to be capable of burst-power that would be useful for Earth-to-Orbit. And that's because they have high surface area for greater coupling. So likewise, couldn't that similarly be exploited for an N-battery? Just grind your N-fuel into particles and encapsulate these with the nanomaterial for faster energy conversion and higher power output.

    I'm also wondering if buckyballs wouldn't be better than nanotubes?
    Because even in chemical fuel cell research, buckyballs have been found to provide more useful surface area and bulk porosity. In the case of an N-battery, you could encapsulate the N-fuel directly inside the buckyballs/buckyonions, for most interfacial coupling and better power output.

  8. Mar 31, 2008 #7
    Well, the energy source could be Plutonium-238 as the nuclear fuel.

    Since your electric current is your energy transfer medium, unlike a fluid in regular reactor, then this might be safer to work with, so that you don't run the risk of mechanical breakdown.

    Fluids require pumps, etc, but electric current allows everything to be solid-state.

    And current would be useful for plasma propulsion (eg. VASIMR or MHD/MAD)

    But again, I'd wonder if the power output could be concentrated in some kind of burst mode, as with a particle-bed reactor. That would be useful for a launch vehicle, which basically needs 10 min worth of high-power output.
  9. Apr 1, 2008 #8
    I remember that when particle-bed reactors were being researched for propulsion purposes, they were found to suffer from fluid-flow instabilities. I guess that makes sense, since maximizing your surface area for maximum interfacial contact and heat transfer is also going to maximize your frictional surface contact and associated turbulence, etc.

    Isn't that why they say that having one big rocket engine is better than having many little ones strapped together? Not just from the perspective of minimizing the number of failure points, but also in regards to efficiency. One big rocket is more efficient than many little ones, because then you've got reduced surface-to-exhaust contact, etc.

    So if particle-bed reactors are handicapped with respect to fluid flow, despite their "promisingly high power output, then maybe this "radiovoltaic"/alphavoltaic mechanism of direct energy conversion is the best way to go. Maybe that's the best way to overcome the fluid flow handicap -- don't use fluid flow and heat transfer to propel your propellant exhaust. Instead, harvest the electricity, and use that to energize a plasma, for an electrically-propelled rocket.

    Use this alphavoltaic method to power a VASIMR-style rocket.

  10. Apr 1, 2008 #9
    The fact that these former LANL researchers are part of some tiny startup LAVM Inc makes me a little leery. However, a lot of people have been pursuing research in Alphavoltaics for quite some time, and even Betavoltaic batteries are due to hit the market soon.







    I was just wondering if this radiovoltaic method of energy-extraction could be more efficient than conventional thermal reactors. Perhaps it could even help an accelerator-driven system move past break-even. Hey, at least it's not limited by Carnot efficiency, and you don't have to go through a lot of energy conversion steps to get to electric power.

    If we can justify researching chemical fuel cells for their superior efficiency over thermal combustion engines, then why can't we justify research into radiovoltaic energy conversion, if it has superior efficiency over nuclear thermal reactors?
    Last edited: Apr 1, 2008
  11. Apr 2, 2008 #10


    User Avatar
    Gold Member

    Where did you get the information that this was based on alphas? There's no mention of radiation species in the OP links.
  12. Apr 2, 2008 #11


    User Avatar
    Science Advisor
    Gold Member

    So here's an interesting idea:

    Depending on what kind of radiation these devices are made to absorb, they could also perhaps be used as a lighter-weight alternative to tungsten, lead, or water for radiation shielding on a spacecraft... If a specifically designed multi-layer surface was able to absorb several kinds of radiation and output electricity, not only would the radiation shield provide extra power for the ship, it would protect the occupants of the ship in the process...
  13. Apr 2, 2008 #12


    User Avatar
    Gold Member

    Without going to check I believe only the density of the shielding/converter is relevant for protection from a given particle radiation. That is, whether the particle impact creates an electron hole pair or the impact merely transfers energy to the shield in the form of heat makes little difference; its the odds of passing through the lattice that determine the level of protection.
  14. Apr 2, 2008 #13
    Well, it's about how much energy is possessed by any particle emerging from the shield. Certainly if a radiation particle is passing its energy into an electron-hole pair, or passing its energy as heat, it means that it's interacting with the shield material, and will come out with significantly less energy, if it comes out at all.

    The reason why I mentioned alphavoltaics, is that they provide the most kinetic energy, and have the least likelihood of passing through a shield. Those are the desirable characteristics for radiative energy capture. You don't want gamma-rays or neutrons, which are hard to block. Of course, it's been shown in the past that alpha radiation will quickly destroy the structural integrity of any ordinary semiconductor that's not radiation-hardened. That's why alpha-voltaics hasn't taken off so far.

    You need some exceptional new semiconductor with extraordinary properties to be able to withstand the alpha bombardment for any reasonable length of time. I'm wondering why this new proposed nanomaterial is able to do it? Like I said, could it be due to electromigration, caused by the electric current running through the nanotubes? Are the nanotubes sufficiently transparent to the alpha particles, so that they don't get turned into swiss-cheese under alpha bombardment?

    Like I said, liquid semiconductors like liquid Gallium have been researched as well, since with a liquid there's no concern about losing structural integrity. But it's not clear to me what the results of that research has been.

    The reason that betavoltaic batteries are now making it to market ahead of any other radiovoltaic material, is that beta-battery devices are harvesting electrons from beta-decay, which are much milder to intercept. That of course means the power they supply is much lower too, although long-lasting.

    But anyhow, what if radiovoltaics could be considered a superior alternative to the traditional harvesting of thermal energy from nuclear reactors? It's just like comparing a fuel cell with a combustion engine -- the fuel cell is more efficient, although it may be harder to engineer. Or comparing a photovoltaic solar cell with a solar thermal collector. It's easy to get anything warmed up by sunlight, but it's a little harder to engineer a solar panel that exploits the photovoltaic effect. And yet society is now increasingly looking towards fuel cells and photovoltaic devices, and not merely satisfied with the old thermal stuff.

    So could it now be time for radiovoltaics to emerge as well, as a superior alternative to nuclear thermal reactors?

    Moving beyond the phenomenon of mere nuclear decay -- what about FISSION?
    Can a radiovoltaic device be engineered to harness the energy released from a fission reaction, rather than a mere decay process?
    Fission is at least CONTROLLABLE, whereas decay merely occurs at some steady rate (at least in a bulk material)

    So it would be preferable to harness energy from a CONTROLLABLE PROCESS LIKE FISSION, rather than from some decay process whose rate of power output we're stuck with. If you were trying to power a rocket, harnessing a decay process might be akin to burning solid fuel (which can't be throttled), whereas harnessing energy from a CONTROLLED FISSION process would be like burning liquid fuel (which can be throttled).

    Sure, you could argue that one could select the appropriate starting-fuel material having the appropriate decay chain, such that it could provide you with roughly the desired power output for your application (like launching a rocket from Earth to Orbit). But that's not the same level of control as a throttleable process, which is what you'd really like.

    I'm wondering if diamond or a diamondoid/diamond-like material could do the job of radiovoltaically harvesting the radiation energy from a fission reaction, with higher efficiency than a thermal nuclear reactor, and with reduced shield weight in the bargain?
  15. Apr 2, 2008 #14
    I was wondering about buckyballs. Buckyballs are supposed to have tremendous structural integrity, as well as good conductivity. Nanotubes are strongest under tension, while buckyballs are good under impact.

    They can also behave like a fluid, which means they could circulate around the nuclear fuel elements and distribute the damage among themselves, rather than sitting in one place where the damage can accumulate more quickly for some. Furthermore, electric current from the radiovoltaic effect could assist in repairing the buckyballs via electromigration (especially if you have a mix of graphite circulating with your buckyballs). The heat from the nuclear fuel elements could also help.

    The buckyballs would have more surface area than mere nanotubes. I'm wondering if the buckyballs could be directly mixed with the nuclear fuel, for maximum contact. Perhaps the gold could be encapsulated inside the buckyballs. Or else maybe the fuel could be encapsulated inside the buckyballs, with the gold on the outside. I'm not sure which way would make more sense.

    Last edited: Apr 2, 2008
  16. Apr 2, 2008 #15
  17. Apr 2, 2008 #16
    Read this:


    Apparently, buckyballs can speed up the natural beta-decay rate of an atom.
    It seems like the higher electron density inside the buckyball does this.

    If you can speed up the natural decay rate, you can get more power output out of the decay process.
    Last edited: Apr 2, 2008
  18. Apr 3, 2008 #17
    This would only be valid however for nuclei that decay via electron capture.
  19. Apr 3, 2008 #18

    Read this:


    Alpha-radiation can destroy a semiconductor's structural integrity and performance within hours. And so if you're going to use a solid-state semiconductor, you need some way for it to "heal" or repair itself in situ.

    That's where electromigration can some in. High current flow through the nanotubes (or buckyballs) could repair any holes created by the alpha-bombardment. This will maintain their structural integrity, and allow them to withstand the alpha-radiation for longer duration and higher intensity.

    I'm even wondering if electromigration could help a material withstand all the various types of ionizing/destructive emissions from a fission reaction, such as fission fragments, alphas, betas, neutrons, and gamma rays.

    If holes get punched into the graphene, just let the electromigration repair them.
    Last edited: Apr 3, 2008
  20. Apr 3, 2008 #19
    Another thing, suppose you treated your fission radiation-absorbing bandgap material as disposable, because you only needed its power temporarily, such as for powering a rocket.

    Then even if the radiovoltaic bandgap material was destroyed by the fission radiation, you wouldn't mind, just as long as it performed well for the first several minutes that you needed it to power a rocket.
  21. Apr 3, 2008 #20
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Nanomaterial Converts Radiation to Electricity
  1. Radiative Waste (Replies: 2)

  2. Radiation device (Replies: 12)

  3. Radiation Filtering (Replies: 1)