I've only skimmed the patents, written in patentese, and so I could be way off base. But, damn, reading through this stuff gave me a geekout. Clue one as to what they might have done is the material’s flexibility. The back electrode is metal, nothing too interesting there, but the front electrode has to be a) flexible and b) clear. AKA conducting polymers. Behold patent 6,936,761 "Transparent electrode, optoelectronic apparatus and devices". Basically it describes a thin film of electrically conductive wires embedded in conductive polymer. There's quite a list of polymers in that patent, so let's talk about plastics that conduct electricity instead. Plastics are generally insulators, not conductors. That's because the electrons in plastics are tightly bound. If you want conduction, you need electrons that are free to move around. (As an aside - and herein lies the trick - electrons that are happy to move around are also, by and large, happy to form new chemical bonds. Such as, for the sake of an example, during chemical reactions that make polymers.) Take a look at poly(acetylene). See that alternating pattern of single and double bonds? That's what we want - it's called conjugation, and those double bonds can interact and provide a pathway for electrons to move along. Here's another. So you have a flexible transparent layer to let the light through, and electrically conductive enough to let the system work...- oh, yeah, that bit. Electrons interact with light. They absorb it, and in doing so gain energy. Usually they then re-emit light and move back down to lower energy levels. In photovoltaics (PV) though, rather than lose their energy as light, they are able to use this extra energy to become extra mobile. IF they can get away in time from the, now electron-less, hole that they have left behind. So, you need a conductive material that will capture these happily free electrons and allow them to be put to work. And you need another material on the other side that can handle the job of putting them back, and allow the electric circuit to be completed. That's what you need a transparent electrical conductor for. Another cool bit is that they seem to have used surfactant templating. There is a fairly recent patent covering use of these materials (here's another). Their explanation in the patent basically runs that an electron freed by light has only a short amount of time, and hence a short distance it can travel, before it gets recaptured by the PV material. Effectively what they've done is created a method to make individual particles of PV material that are just small enough to let the electrons out before they can be recaptured. So let's talk about surfactant templating. A really well known material that uses surfactant templating in its manufacture is MCM-41. MCM-41 was originally researched as part of the search for petrochemical catalysts. It's neat stuff. Aaaand, soap. (Bear with me here). Soap is a surfactant, long molecules with water friendly chemical groups on one end (the hydrophilic end) and oil friendly at the other end (the hydrophobic end). Soaps get things clean by allowing oils to be stabilised into water - the oil happy ends of the soap surround the oil particles, while the water happy ends, which are pointing out, are accepted by the water. The oil inside is hidden from the water by the hydrophilic heads of the soap molecules. And the oil can then be washed away. Now what happens if only surfactant is present in the water? The water happy ends don't have a problem, but the oil happy ends do. So the surfactant molecules will first move to the water's surface and stay there, where the oil ends can happily sit near air (making the soap a SURFace ACTive AgeNT) but if more surfactant is added then there is a problem. At high enough concentrations the surfactant molecules start to pool together into structures called micelles. A simple micelle is a sphere, with the hydrophobes inside the sphere and the hydrophiles on the outside. Another shape is a tube, again water haters inside, water lovers to the outside. NOW we have our template. It's the micelle tube. With a little bit of chemistry, silica can be added to the outside of the micelle, forming a hard coating around it. At this point the surfactant is still inside but for the next step the whole thing is put in a furnace and the surfactant is burned out, leaving behind the silica tube. Since the surfactant molecules are all the same length the micelles themselves are very regular in diameter (not so much in length) and the templated silica inherits that property. It looks to me like the same _sort_ of chemistry is being exploited to make a nanoscale PV material, of the right size to capture light and release electrons. Judging by the patents they do some post processing to introduce the right mixture of materials into the system to make the templated material photoactive. I could be completely off base here. But this is fun stuff and some nifty chemistry seeing the light of day.