It kinda depends on what your exact goal is, but there's a few different ways to make a solar cell. http://www.youtube.com/watch?v=FrdUlPM4x0Y&feature=relatedThis is a basic setup for a basic solar cell and it's a good walk through for the basic process. After you watch the vid, there's ways that you can change the basic idea. For instance, if you use polyaniline, instead of indium-tin-oxide, you can drop the price of making the solar cell by alot. Wiki Polyaniline if you're not familar with it as a material. it's interesting, and can be made in small quantities with materials from the hardware store. After that... I will say that the bottom panel can be opaque, which might open up some materials as options for future developement by DIYers (*cough* aluminum foil or aluminated mylar *cough*).Indium-tin-oxide is expensive, and using a less expensive substrate to produce the solar cell drops cost by alot. After dropping the cost, i'd focus on increasing efficiency. The best idea i've seen involving increased efficiency involves consideration that the thin layers of material are still 3 dimensional, even though we, on a large scale, only think of them as two dimension (that is, people who design solar panels think of the solar collector as a plane having only x and y coordinates, and the not as a cubed object.).Prisms, crystal structure, and composite materials are all facets that can come into play, when you start looking at on this level. I'm kind a saving up so i can buy some Bizmuth... i'm pretty sure some of it's electronic characteristics can greatly increase solar panel efficiency.