Or, I suppose you could park a Stirling engine on it and generate power from the thermal gradient it must be producing.
This same group is also trying to solve a related problem, which is how to keep solar panels cool, since they are less efficient when they heat up. They have designed coatings that are transparent to sunlight and dump heat to space ( http://www.opticsinfobase.org/optica/fulltext.cfm?uri=optica... )
That said, the material is specified to have a cooling capacity of ~40Wm^-2. You normally assume something like ~800-1000Wm^-2 for incident solar radiation, so I'm not sure how economical using something like this with a solar array would be.
Really cool though.
Well, a sphere of expanding energy, anyway. Just like the shell of TV broadcast energy that leaves the planet. Also, we rotate and move in space, so less of a sphere and more like a windy tube.
http://en.wikipedia.org/wiki/File:Atmosfaerisk_spredning.png
As an example. Visible light is that little bit all the way to the left. That entire big chunk on the right is the thermal window that they're talking about.
However you wouldn't have to worry about optimizing these for the atmospheric IR window, as you do on earth, which is the advance claimed in this paper.
[0]: http://www.nature.com/news/nature-makes-all-articles-free-to...
