[coreytaylor]

Solar radiation spreads over a large range of wave lengths, although its quite concentrated over the visible spectrum, there’s a big distance from red (lowest energy visible) to violet (highest energy visible photons).

The most efficient solar panels use up to 3 junctions, each optimized for absorbing one specific wave length range, but still a lot of radiation can’t be absorbed at all (infra red+below and ultra violet+low x ray and radiation in the middle of each junction optimal wavelength) and a lot of the radiation each junction can absorb at all has a lot of waste that never makes it to electricity.

And such panels above 25% efficiency are already very expensive (used pretty much only for satellites and other applications where weight and size is critical like mobile solar panels that US military carries around to generate their own electricity instead of carrying diesel and generators). Achieving 5 or 10% more efficiency isn’t worth it if it means another 30–50% cost. Even in such applications its usually more cost effective to use more panels than an even more expensive one.

Theoretically it should be possible to get to 60% efficiency eventually, but that will likely only be achieved if it comes with a very modest increase in price/watts.

In essence the high efficiency solar panels are good enough for satellites already. The trends is towards more and more solar panels in them, so they can be used for more transponders and to power up ion thrusters for lower mass transit from GTO to GEO.

Mankind would benefit a lot more from 25–30% efficient affordable solar panels than a 60% super duper expensive ones.