(Note: the following is one of various posts that were copied here from the Moonwards.com forum for the sake of preserving the early days of the project. It was originally posted there on Oct 5th, 2015)
So here we have a convex parabolic mirror focusing light on a secondary concave parabolic mirror, which passes it to a third mirror as one semi-coherent beam, and that mirror sends it to the big crucible filled with regolith placed to one side. In the proportions shown here, the resulting beam of sunlight is concentrated by a factor of nearly 300. That might be too much. Heat dynamics is way beyond me, so the thermal concentration here may be fine, or it might melt everything even if it is made of a tungsten skin over a graphite interior and has a radiator on top (currently not shown). My guess is the 2nd option is more likely. The concentration factor can of course be adjusted to whatever works best. Still, the higher it can be pushed, the better, right?
The mirrors themselves could be a thick layer of aluminum to deal with the constant loss of aluminum from the surface due to its low vapor pressure. By circulating a liquid (a molten salt?) through channels in the graphite the heat could be used to turn a micro-turbine before being cooled in the radiator. Thus some use would made of it. Perhaps not worth the added complexity though. There is a microwave transmitter attached to the counterweight near the pivot of the whole thing in this model, which went with the idea the microturbine idea could be worth it. It passes the power generated to the crane for distribution.
The crucible holding the lava, by contrast, would be surrounded by reflective insulation (except where the light beam hits it) to limit heat loss. Its whole interior volume needs to reach 1500 Celsius at least to produce lava, and really the target is 2500 Celsius or more. At that temperature, oxygen can be produced in volume.
The idea of this configuration is to keep a volume of lava boiling away for an extended period of time. I think that is what is needed to have a process like oxygen production be effective. A cylinder that is say 1 unit wide by three high might also separate into layers over the night as its contents slowly cooled, right? Which would mean its different constituents could be separated and then processed further.