Today our prof told us (with a completely straight face):
"People always worry about what happens when you put a solar cell out into the sun."
And they do. Most types of solar cells work perfectly fine in the dark, i.e. they produce no current at all, exactly what you would predict, but put them into the sun and they'll be up to all kinds of shenanigans. They'll degrade, they'll heat up and be much less efficient than predicted, etc. etc. Really, if you want your solar cells to last, keep them in a cool, dark place!
The problem is of course that the semiconductor of a classical cell will heat up, which will change the characteristics and kill efficiency. UV will do all sorts of evil to your cell too.
Dye-sensitised solar cells (invented by Prof. Grätzel here at the EPFL!) don't suffer from heat-inefficiencies, and offer the additional advantage of working in low-light conditions as well (Both feats are due to the fundamentally different approach and quantum physics this is actually kinda cool if you are into this kind of stuff). Unfortunately, they degrade under continuous operation, and since they utilise a liquid electrolyte, they suffer a more destructive reaction to temperature extremes: The electrolyte will either freeze (and crack the cell) or expand (and crack the cell). Still, they might get that under control, or maybe we can switch to a solid electrolyte, as reported by Prof. Grätzel in Nature Materials last year. Although then you loose the advantage of flexible cells...