Rectennas and 2nd law of thermodynamics

[theDoc]

Rectennas are diodes attached to antennas. With current technology they can convert EM waves with frequency up to $$\approx$$ 1 THz to electricity.
Imagine an isolated room at 300$$^{o}$$ K containing a black object and a rectenna. By Planck law, the black object radiates about 450W/m$$^{2}$$, of which about 1/10,000 is of frequency less than 1 THz. Couldn't this .045W/m$$^{2}$$ be captured from the waste heat, in contradiction with the 2nd law of thermodynamics? For e.g. use this electricity to heat another room and create a temperature differential

 

[alxm]

That's basically a variant of http://en.wikipedia.org/wiki/Maxwell%27s_demon" variant.

 

[theDoc]

I don't see how Feynman's argument about the ratchet works in this case.
This is a very simple experiment: at room temperature, place a 1 sq. meter sheet of black material (for e.g. http://www.pnas.org/content/106/15/6044.full) on top of a 1 sq. meter sheet of THz rectennas (like in http://www.techbriefs.com/component/content/article/2165). Wouldn't you be able to collect .045W of electricity?

 

[berkeman]

I don't see how Feynman's argument about the ratchet works in this case.
This is a very simple experiment: at room temperature, place a 1 sq. meter sheet of black material (for e.g. http://www.pnas.org/content/106/15/6044.full) on top of a 1 sq. meter sheet of THz rectennas (like in http://www.techbriefs.com/component/content/article/2165). Wouldn't you be able to collect .045W of electricity?

I'm no help on the thermo part of the question, but keep in mind that the antenna terminal voltage has to exceed the forward voltage drop of the rectenna diode. At 1THz, the terminal voltage for any modest fields will be quite small. I should probably do the calculation, but I don't think it's near a diode drop.

 

[cesiumfrog]

I was thinking the mechanism of failure would be that as the diode warms to the temperature of the radiation, spontaneous electron-hole pair creation would be at a rate that cancels out the meagre net current. I don't know whether this is equivalent to what berkeman is saying?

 

[berkeman]

I was thinking the mechanism of failure would be that as the diode warms to the temperature of the radiation, spontaneous electron-hole pair creation would be at a rate that cancels out the meagre net current. I don't know whether this is equivalent to what berkeman is saying?

Interesting thought, but different from what I was referring to.

For a given field strength, say E = 1V/m, the terminal voltage at a resonant antenna is on the order of the size of the antenna. For a 1m dipole antenna at resonance, a 1V/m E field will give about a 1V AC receive signal (I'm using order-of-magnitude here). As the wavelength decreases for the same E field strength, the receive voltage ratios down with the wavelength. So a resonant antenna in a 1V/m field that is 0.1m long will only give you about 0.1V AC at its output terminals.

http://www.ipllc.cc/RFID%20and%20FCC%20Part%2015.pdf

So you can see where you end up at the frequencies asked about in this thread -- very small antenna voltages, probably too small to rectify. The only thing helping proposed future solar energy rectenna arrays (which have other issues), is that they try to concentrate the radiation quite a bit, to get high field strengths that could be rectified by exotic light-frequency diode structures.

 

[cesiumfrog]

Can't you counter that by making the antenna arbitrarily large?

 

[Gordianus]

We don't have to consider real diodes that show voltage drop and so on. The rectenna is a beautiful sibling of Feynman's ratchet and pawl. Remember how he stressed the necessity of a dissipative process that would prevent the the pawl form bouncing back and forth and how this process would heat up the whell until its temperature rises too much and the brownian motion might allow the axle to turn backwards. The rectenna would presummably feed a resistive load. This resistance would heat up and "save" the second law of thermodynamics

 

[berkeman]

Can't you counter that by making the antenna arbitrarily large?

Not to increase the terminal voltage. There's no way that I know of to "series" connect antennas to increase their receive voltage output. You have to boost the receive E-field to do that.

I guess I should be careful about not going OT with this antenna discussion. I think the main focus of the thread is really the thermodynamics of the question. Maybe if there are further comments/questions about the antenna aspect of the question, should we handle that via PM?

 

[alxm]

I don't see how Feynman's argument about the ratchet works in this case.

It's essentially the same argument, except you're using a diode instead of a ratchet and pawl, and thermal radiation instead of mechanical heat (which are in equilibrium).

You have no net motion of the ratchet and you have no net current across the diode, when it's in thermal equilibrium.