Actually you are understating the difficulty of transmitting that power.

There are several big problems. One is real world rectennas are not that efficient. The best best lab condition ones with lower power are 90% but high power ones are quite a bit less, sometimes up to 75 percent or so.

The other massive loss is transmission loss. The basic formula is 32.45 + 20log(d) + 20log(f). Using the 1.32 TW estimate in this post: http://science.slashdot.org/co... and a rough 24ghz given in this page http://www.propagation.gatech.... we would have around 231 db. Considering you are starting with 121 dBW of power, you are left with -110dBW of power on earth not accounting for antenna gains.

Since a watt is 0dBW, we need 55db of dish gain on either side to get 1 watt and a gain of 115 to get the input power generation. At 24 GHz this is a 3 KILOMETER dish on either end. The largest parabolic on earth now is 0.3 km, so ten times that. Plus it would have to always point at the moon (how? No idea).

Plus You would lose your 50% or more since obviously the whole equator isn't facing the sun. Then the 50-70 from your rectenna loses.

I imagine a bunch of other stuff from other losses I havent taken into account (Convert to AC, line loss, whatever else) and I cant imagine this being feasable.