Electricity from beta decay?

[blakat1313]

I didn't find a thread on this when I searched, so I don't think it's been done before.

I thought of something a while back and I'm wondering if it could work. When an element goes through beta decay, it produces a stream of electrons, right? Would it be possible for this kind of element to be used as a generator for electricity? I can't find any records of it being done before, but I could be reading the wrong articles or searching the wrong terms.

[Charlie!]

Well, it also causes the creation of a proton, so I'm going to guess that any flow of charge is minimal.

Er, unless you very quickly separate the electron from the source... maybe by having it jump across a vacuum? Bonus: use the electron source as your positive terminal. This method would be finicky, limited by surface area, and probably a net energy loss (harvesting the source). But it would be cool, I guess

[Matterwave1]

I think there'd be too many problems with this.

One, radiation is isotropic (usually), so you'd hafta find a way to get the electrons to move in the direction of the current.

Two, radiation is random (as far as I know, tho I haven't studied QCD), so you'd get fluctuations in current (tho this may be small if you have a large enough sample).

Three, the emitted electrons are high energy electrons.

I think there's others, but i can't think of any atm.

[blakat1313]

That would explain why I haven't found anything on this then. Still, you could control the direction of the charge by surrounding it in a sphere of conductive material with one wire off of it. Have a few spheres leading to batteries and the fluctuations could be helped. The proton does throw a monkey wrench into the works, though. This is the first I'd heard of the proton though, can you show me a link or something? This sounds different from what I learned and I'm curious how it works.

[Minerva]

You can in fact make usable electrical current.

http://www.rochester.edu/news/show.php?id=2154

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

The proton does throw a monkey wrench into the works, though. This is the first I'd heard of the proton though, can you show me a link or something? This sounds different from what I learned and I'm curious how it works.

Basically, beta decay is where a neutron in the nucleus turns into an electron plus a proton plus an electron antineutrino. The electron, or beta particle, is shot out of the nucleus, whilst the proton remains in the nucleus, making the nucleus a different nucleus with +1 atomic number. Whilst electric charge is conserved overall, there is a brief creation of a net positive charge around the vicinity of the nucleus.

[blakat1313]

Thanks for clearing that up, I was thinking that the proton was ejected along with the electron and that sounded off to me. Why do these things need to have fancy names? I searched "electricity from beta decay" and didn't get anything, I wouldn't have thought of betavoltaics in years. I guess it makes sense in retrospect.

[MarvinM]

Invented in the 50's?

Put's me in mind of Strutt's "radium clock".

http://www.lateralscience.co.uk/radium/strutt.html

[meat.paste]

You absolutely can generate current with beta decay. In the gas chromatography (GC) world, a Ni-63 beta emitter is routinely used to generate ~10 nA of current. The actual beta decay current is less, but the emitted electrons have so much energy that they strip off electrons in the surrounding gas as they thermalize.

[frezik]

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

Alpha decay is prefered, since its easier to shield. Beta decay can produce a lot of x-ray/gamma ray emissions as a side effect.

Various deep space probes were built with RTGs as their main power source. The Solviets built a bunch of lighthouses with them, though poor documentation means that they don't know where all of them are anymore.

[Monty40xi]

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

Alpha decay is prefered, since its easier to shield. Beta decay can produce a lot of x-ray/gamma ray emissions as a side effect.

Various deep space probes were built with RTGs as their main power source. The Solviets built a bunch of lighthouses with them, though poor documentation means that they don't know where all of them are anymore.

That's a little different. RTG's are powered by the heat of radioactive decay, not by the charge of the radiated particles.

[MarvinM]

meat.paste,

Isn't that much used for the detection of ppt's of chloro/bromo carbons with ring systems in GC? Correct me if I'm wrong but a voltage is also involved I think which creates the current and the beta source just creates the charge carriers in the output gas. It's a really cunning method and extraordinarily sensitive for some types of compound from what I remember.

[meat.paste]

meat.paste,

Isn't that much used for the detection of ppt's of chloro/bromo carbons with ring systems in GC? Correct me if I'm wrong but a voltage is also involved I think which creates the current and the beta source just creates the charge carriers in the output gas. It's a really cunning method and extraordinarily sensitive for some types of compound from what I remember.

It is used for the selective detection of halogenated species at very low levels. It doesn't have to be a ring system, necessarily. It is very sensitive, although the absolute detection limits will vary based on the number of halogens and the exact kind. O, S, and N can also absorb an electron, but to a lesser degree than the F, Cl, Br, or I. Something like bromoform (CHBr₃) is likely to be detectable to near 1 part per trillion (1 ng / kg).

The voltages present in the detector are for biasing, not for current production. Basically, once the ion has been made, it needs to get to a collector electrode before being swept away by the GC carrier gas flow.

[Teppic]

I didn't find a thread on this when I searched, so I don't think it's been done before.

I thought of something a while back and I'm wondering if it could work. When an element goes through beta decay, it produces a stream of electrons, right? Would it be possible for this kind of element to be used as a generator for electricity? I can't find any records of it being done before, but I could be reading the wrong articles or searching the wrong terms.

You are quite correct. It technically could be made into a current source. However;

1. The gradient (PD) would be very very small; the energy of the particle after it overcomes the coloumb barrier of the nucleus is eV at best.
2. Its rather chaotic. Literally. Although any small lump of beta emitter would technically create a current (flow of charged particles) it would be like filling a water balloon with a waterfall.
3. It would be impossible to switch off. And a combustion hazard
4. Batteries do it without being carcenogenic
5. Very rarely do you get pure beta emitters. Expensive. And deadly

I do see your point, it would be direct conversion from energy source to energy carrier rather than go through the steam turbine route. As Bill Bryson points out (and despite particle physics being a speciality in my degree I dont know why it is so) nuclear fission only releases 2% of an atoms available energy.

What is slightly more interesting, to me from when I was taught it, is the inverse of beta decay (almost); quantum tunneling of electrons in semiconductors. Which form quantum currents. Such a low probability wave function allows tunneling through solid objects and even a steady current which allows me to type this right now.

[Teppic]

Well, it also causes the creation of a proton, so I'm going to guess that any flow of charge is minimal.

Actually in a lab environment beta sources have to be earthed. Indeed I remember a field equation question about the charge build up on a beta source.

[MarvinM]

The voltages present in the detector are for biasing, not for current production. Basically, once the ion has been made, it needs to get to a collector electrode before being swept away by the GC carrier gas flow.

Biasing for the purposes of sweeping out charge carriers means current production. The energy in the current comes from the biasing field. It's still an utterly awesome method.

Teppic,

Word to the wise, posting under alcohol/caffeine/nitrous oxide/corner shop sweets, not advised, best do some hasty editing to your post before someone rips it apart. Particularly the particle physics at degree level bit, someone is going to carve you a new a hole.

[meat.paste]

Biasing for the purposes of sweeping out charge carriers means current production. The energy in the current comes from the biasing field. It's still an utterly awesome method.

Ultimately, any current source requires a return flow. Even without the bias current, the electrons will flow. Certainly, there is some additional energy that arrives from the ions falling through the potential field. However, the dominant amount of energy is coming from the 67 keV electrons from the beta decay. I agree that it is a pretty awesome idea. The output current is very stable for a very long period of time (yay physics!)

[Monty40xi]

I guess the question is, can you get more energy from harnessing the electrical charge than you can from using the particle for heat, as in the RTG?

[meat.paste]

I guess the question is, can you get more energy from harnessing the electrical charge than you can from using the particle for heat, as in the RTG?

I think it depends on the specific isotopes involved and the efficiency of the thermoelectric conversion. A 10 mCi beta emitter will directly produce 3.7x10⁸ e^-/s = 59 pA. If the beta emitter is Ni-63, then each electron is carrying 67 keV. If you extracted all the energy from the electrons, then this would be 4 microwatts. A 10 mCi source of Ni-63 should mass 0.16 mg, if I've done my math correctly. So, Ni-63 could generate a maximum of 24.5 mW/g. Pu-238 has a decay rate of 17300 mCi/g, so a 10 mCi source contains 0.578 mg Pu, or 0.656 mg PuO₂. Because the decay mode for this isotope spews out 5592 keV alpha particles, the maximum available heat from the conversion of kinetic energy would be 331 microwatts or 505 mW/g. The efficiency of the heat to electricity conversion is ~5%, so the Pu-238 would make about 25 mW of electricity/g PuO₂. Assuming the beta decay to electricity efficiency is ~50% (seems reasonable since the radiation is already current, the electrodes could be shaped to capture a significant fraction of the emitted particles, and the emitted electrons can be used to generate secondary electrons in a gas as the voltage drops), then Ni-63 would generate about 12.3 mW of electricity/g Ni-63. These are pretty close to each other, which gives me some confidence that there would be an isotope that would generate more electricity per gram through beta decay than the thermoelectric conversion of Pu-238.

A quick "reasonable-ness" check - the 10 mCi Ni-63 actually generates about 10 nA of current in a gas filled chamber (knocking additional electrons off of the gas as it slows down). Assuming the additional electrons have ~10eV of energy, this means the Ni-63 source would generate ~0.1 microwatts. The same decay rate of Pu-238 generates 16.4 microwatts, so the RTG has a 40-fold increase in electricity production on a gram basis.

[kb5uew]

Here is a thought:
The direction and speed of Electrons can be easily controlled with a potential such as in vacuum tubes. Any stream of moving electrons produces a magnetic field. You harness the changing magnetic field to produce AC electrical output such as in a transformer. Can these known effects be combined to produce more wattage magnetically than just harnessing the electrons or heat?

[idobox]

The easiest way to produce electricity from a beta emitter would be to put it at the center of a hollow metallic sphere. Beta particles will charge the outer electrode negatively, while the inner will charge positively.
In the beginning, most of the energy will be lost as X-Rays, but as the charge builds up, it will slow down the electrons, until the potential is equal to the energy of individual particles, and almost all the energy of the beta particles will be converted to electricity.
Proponents of the fusor and polywell want to use this with alpha particles. Of course, since the source would be a low density plasma, instead of a solid electrode, much less energy would be lost to collisions.

[Sandor]

Some of this discussion reminds me of the idea for direct conversion of energy to electricty with aneutronic fusion. Unfotunately, I know nothing more about it than what appears in the Wiki article.

[w.eckhardt]

I only now saw this post.

Last year in the Dutch final nationwide schoolexams in physics, there was actually a question about one such system.

A small nuclear battery was proposed, consisting of a Ni-63 source.
The radiaton will charge an small copper plate, and that will attract to the original source (which now has an opposing charge). On contact, the charges will cancel each other out, and the system returns to its starting state.

This constant movement is then converted in electricity by means of a piezo-electric material.

This battery will not generate much power, but as a long-term device, it could very well be used.

Direct link - it's on page 4. Warning: The pictures are nice to look at, but you have to be able to read Dutch.
http://static.examenblad.nl/9336111/d/ex2011/ha-1023-a-11-1-o.pdf

[idobox]

So it's using electricity to move something to generate electricity? Why not directly convert the 17 kV DC charge to a useful current? (I can't speak Dutch)

[PM 2Ring]

So it's using electricity to move something to generate electricity? Why not directly convert the 17 kV DC charge to a useful current? (I can't speak Dutch)

Well, there are losses when converting high voltage, very low current DC to a more useful voltage, and the spherically symmetrical geometry makes completing the circuit a bit awkward, since your return line to the positive pole gets in the way of the emitted beta particles, blocking their path to the negative pole.

Wikipedia has an article in English on various atomic battery technologies, and if you read the full article on the Radioisotope piezoelectric generator you'll find a link to a Cornell news item on this cantilever-based device. From my reading, it appears that the main advantage of this device is its high impedance, since very low current EMF isn't much fun to work with.

[idobox]

DC-DC downconversion is quite efficient, but it is not often done with kV. It must be quite difficult to find the right transistors.