# Nuclear battery

1. Aug 14, 2006

### {~}

I was going to post this in the nuclear engineering forum but my question is really more electrical.

A beta emiter is a radioactive substance which radiates electrons (beta particals) in all directions.

My question is, is there a way to generate electricity from electrons traveling away from a spherical source?

2. Aug 14, 2006

### dlgoff

Most beta emitters (well the ones I've worked with) are not very energetic. Like C14; the particles can be stopped by a couple sheets of paper. But I quess there's a little energy you could gain from. I don't know how you would make a battery though.

Regards

3. Aug 14, 2006

### {~}

well wha if it were a strong source? Isn't ther any process in electronics where you might do something like this?

4. Aug 14, 2006

### Staff: Mentor

Put the emitter ball inside a larger concentric sphere, and connect the load resistance between the outer metal sphere and the inner beta emitter. Seems like it would work.

5. Aug 14, 2006

### chroot

Staff Emeritus
Problem 1:

A metric ton (1000 kg) of carbon-14 has about $6 \cdot 10^{29}$ atoms. According to http://en.wikipedia.org/wiki/Carbon_14, Carbon-14 has a specific activity of 14 disintegrations per minute (dpm) per gram carbon.

Thus, a metric ton of carbon-14 undergoes about 233,000 disintegrations per second.

A typical hair dryer consumes 15 amperes of current at 110 volts. This means that approximately 200,000,000,000,000,000,000 electrons are travelling through the hair dryer every second.

If you wanted to power your hair dryer with the electrons ejected by beta decay, you'd need something on the order of 30 trillion tons of carbon-14. Doesn't sound very efficient, does it?

Problem 2

As the electrons are ejected from the atoms of a beta-radioactive substance, the remaining atoms become more and more positively charged. It would be quite difficult to continue pulling electrons from them indefinitely.

- Warren

6. Aug 14, 2006

### NoTime

I can't see this as any normal battery, because you arn't generating any holes for the electrons to move back into.
OTOH if you simply put a sphere around the source it seems like the sphere should eventually generate a negative charge with respect to ground(in this case literally ground)

Problem is that 1 amps worth of current for a second represents something like 6 X10^18 electrons. That is a lot of decay events. Using it as some sort of power source seems to be shy a few orders of magnitude.

7. Aug 14, 2006

### NoTime

Do they?
If C14 decays to C13 arn't the number of protons the same. The electron shels would stay the same with the same occupancy.

8. Aug 14, 2006

### chroot

Staff Emeritus
How in the world is carbon-14 going to decay to carbon-13? Are you suggesting that somehow a neutron just "jumps" out of the nucleus?

Perhaps you do not understand what beta decay is: the decay of a neutron into a proton and an electron (and an electron antineutrino). Carbon-14 decays to Nitrogen-14 via beta decay.

- Warren

9. Aug 14, 2006

### chroot

Staff Emeritus
This is not true -- this would violate the conservation of charge, which does not happen in any particle decay.

- Warren

10. Aug 14, 2006

### NoTime

Fair enough.
My knowledge of decay paths leaves something to be desired.
OTOH this decay path generates a hole for an electron to move back into as in berkeman's post.

PS: Wouldn't that have to be N-13 and not 14?

11. Aug 14, 2006

### chroot

Staff Emeritus
Yes, you could conceivably create a circuit, allowing the ejected electrons to move through a circuit before coming back to the sample. (I thus retract my "Problem 2.")

And no, the decay product is definitely nitrogen-14. Carbon-14 has 6 protons and 8 nuetrons. If one of those neutrons decays into a proton (+ electron, etc.) the resulting atom has 7 protons and 7 neutrons, and is nitrogen-14.

- Warren

12. Aug 14, 2006

### NoTime

What's the basis for the nuclear batteries used on some spacecraft?
Is it something like this with a high rate alpha or beta emitter or is it a thermal process?

13. Aug 14, 2006

### {~}

Spacecraft use a thermal process. The decaying source generates heat which is simply tapped into with thermal couples.

I wasn't expecting to get one electron volt per electron emitted or anything like that. What I was thinking was to try and tap into their kinetic energy using inductance. They do, after all, have moving magnetic fields.

Forget about the whole nuclear thing for minuet. What if you had a magically powered electron gun that could be used to power something else? How would you generate electricity from that?

14. Aug 14, 2006

### NoTime

This was always my impression, but this thread made me realize I didn't actually know how it was done.

I'm thinking you might be worse off this way.
I seem to recall that the ejection speeds are fairly low as such things go and with a mass of 10^-28g or so...
Might do better with an alpha emitter.
Don't know.

Don't see how. The efficiency would always be less than 100%.

15. Aug 14, 2006

### {~}

Does it matter if you don't have perfect efficiency? For the point of this thought experiment the electron gun is powered by something else entirely that we can't tap and aren't controling in any way. The electron gun emits an incoherent beam of electrons. We can't control the gun in any way. The gun could potentialy be a source of energy but the only thing we can manipulate is the beam.

16. Aug 15, 2006

### Staff: Mentor

Radioisotope thermoelectric generators (RTGs) have been used to power spacecraft for decades.

http://en.wikipedia.org/wiki/Systems_Nuclear_Auxiliary_Power_Program
http://www2.jpl.nasa.gov/galileo/messenger/oldmess/RTG.html
http://saturn.jpl.nasa.gov/spacecraft/safety.cfm

:uhh: Magically?!? Well, an electron gun would produce a current of electrons, which IS electricity.

Lower efficiency just means that one has to produce more energy from one source to get a particular amount of electrical energy. For a system of 25% efficiency, one must generate 4 kW of power from a source to get 1 kW of electrical energy. At 33.3% efficiency one would only need to develop 3 kW of power to get 1 kW of electrical energy.

17. Aug 15, 2006

### NoTime

Tapping the beam current seems to be the best way.
Ie: put whatever you want to power in the electron return path.
The beam itself would constitute a DC current, so you would need to steer it in and out of a transformer core or have a pulsed beam.
Also you might get something using the Hall effect.
Edit: Or thermal effects.

Last edited: Aug 15, 2006
18. Aug 15, 2006

### Staff: Mentor

What you are asking about is called "direct conversion", and is part of the road map for fusion power generation. If we could get a clean fusion reaction with direct conversion for the power generation, we would be one happy planet. That's several decades away at least, unfortunately. The best near-term prospects for fusion are still with dirty reactions and thermal conversion...

http://en.wikipedia.org/wiki/Aneutronic_fusion

19. Aug 15, 2006

### {~}

fusion hmm..

The idea to me was sparked by this which looks to me like a gas filled raddiation detector that suposedly generates power.

20. Aug 29, 2006