optics - Light "diode" and 2nd law of thermodynamics

If I had a light "diode" - an object that only allowed light (at least for a range of frequencies) to travel through it in one direction, would this necessarily allow violations of the 2nd Law of Thermodynamics? Or would it be possible to restrict other properties of the device such that it could be okay.

Why I think it should violate 2nd law:

1. 
   

   I can place such devices at temperature T in a bath of thermal radiation at temperature T, such that it forms a box which only lets the light in or out. This would appear to always allow one to create a thermal gradient. Once a thermal gradient is formed, useful energy can be extracted. This seems to allow useful energy to always be extracted from a thermal bath and thus give perpetual motion.

   
   


2. 
   

   Release one such device at temperature T in a bath of thermal radiation at temperature T. Since the photons are absorbed or reflected in one direction, and undisturbed in the other, the device will start moving. It can use a thermal bath to directly generate useful motion.

   
   

Why I think I'm missing something and such a device could potentially co-exist with the 2nd Law:

1. 
   

   The above arguments seem valid if the diode was only 'partially' rectifying as well. However there exist sunglasses which largely reflect on one side, but not from the other side.

   
   


2. 
   

   Researchers have succeeded in using meta-materials to create a "blackhole" for microwaves.

   
   

Can anyone help sort this out?

Answer

Light diodes exist, and yet the 2nd law endures. In general, any perfect ratchet would allow you to violate the 2nd law. Since very few Nobel prizes have been awarded for perpetual motion machines, experimental evidence seems to suggest there is no such thing as a perfect ratchet.

When confronted with any type of device that rectifies the flow of energy, the appropriate question, in my opinion, is 'how is this device imperfect'? In the case of a Faraday isolator, Lord Rayleigh addressed this question a long time ago.

EDIT:
The other reference at the end of the Wikipedia article on Faraday isolators is a pretty nice discussion of this topic.