http://www.bbc.com/future/story/20160705-the-toughest-spaces...
http://www.bbc.com/future/story/20160705-the-toughest-spaces...
Heat dissipation is actually a gigantic problem in space: you can only radiate heat away (and try to reflectively shield against the sun) since you're in hard vacuum which is a fantastic insulator.
An effective radiator is an equally effective absorber, so management of light and shadow is critical.
But it's still sort of moot when you consider atmospheric friction on the way down.
You're not going to get much power out of the gradient. Even if you cooled an object in space to near absolute zero, space probes must necessarily be low mass to get into orbit from Earth. The low mass of the whole object, and the low heat capacity of most space-grade building materials combines to make it a really awful cold sink. You would essentially need to harvest water and ammonia from somewhere outside of a gravity well, use your radiator to cool it off as much as possible, and then drop it in the heat well. That's an awful lot of trouble to go to for the amount of power you will then get out of it.
That's your theoretical maximum output. Assuming the probe doesn't heat up during entry.