One-way mirror ball

**DrGreg** · Nov28-09, 10:19 PM

The principle of a one-way mirror is that from one side, say 90% is reflected and 10% passes through. From the other side, 90% is absorbed and 10% gets through. But it's the same 10% passing through in either direction.

**DaveC426913** · Nov28-09, 11:42 PM

Originally Posted by Integral

Dave,
You can drop the skepticism, he is right on target. In essence all mirrors work that way, all that changes is the percent reflectance.

This even holds true for a plain sheet of glass. At night when you look from a lit room to the outside, you see a reflection of the inside. Now step outside and look through the same window from the dark. You can easily see into the lit interior. That's why we have curtains.

Yes, I am very aware of how windows reflect when in a bright room and transmit when in a dim room. I'm not sugggesting any of the facts about mirrors isn't true. In fact, I know it's all quite true that they do this.

I'm just not 100% certain that the mirror's orientation does not factor in at all.

But if you are certain that one-way mirrors are, in fact, perfectly reversible then I'll take your word for it.

**Phrak** · Nov28-09, 11:52 PM

Unfortunately, mirrors that will pass light in one direction more than the other violate the second law of thermodynamics

--and, by the way, is why Hawking proposed a scheme whereby blackholes would radiate energy.

**DaveC426913** · Nov29-09, 12:12 AM

Originally Posted by Phrak

Unfortunately, mirrors that will pass light in one direction more than the other violate the second law of thermodynamics

Right. I can see that. It's right there with Maxwell's Demon.

If a mirror could preferentially pass light, then you could build a box out of it and make a light (or heat) battery, or a PPM.

**Phrak** · Nov30-09, 12:49 AM

Originally Posted by DaveC426913

Right. I can see that. It's right there with Maxwell's Demon.

If a mirror could preferentially pass light, then you could build a box out of it and make a light (or heat) battery, or a PPM.

Right Dave. I wasn't directing criticism at you, but only making the heuristic argument based on thermodynamics.

(secretly, I rarely pass over the opportunity to find a slim chance at breaking the second law, as it would be the most shattering physics of a century or more.)

Classically, the electromagnetic wave equations are time symmetric. I think this implies mirrors are transmittally symmetric.(?)

But there is still some wiggle room. Light hitting an interface at the Brewster angle is selectively transmitted. To reverse the process, we need to combine reflected and transmitted rays to construct the incident ray in reverse. It's something like getting a proton, an antineutrino and an electron to all arrive at the right place, energy and momentum to make a neutron.

Any thoughts?