I think the amount of toxin would be much smaller than found in, e.g., CdTe panels (8 g/m^2 of Cd), because they may be very thin. But maybe less well contained

I think the amount of toxin would be much smaller than found in, e.g., CdTe panels (8 g/m^2 of Cd), because they may be very thin. But maybe less well contained? (I read claims that a house fire under CdTe panels will not make your neighborhood into a superfund site, but don't know how to evaluate them. Is the Te on one side and glass on the other expected keep the Cd in, even at 1200 C?) There are good reasons to want perovskite cells to be usable. Efficiency can go above 30%, and they are very light and flexible. So, it is great that so many people are working on the problem. I never read anything about them without the person mentioning attempts to improve lifetime. They might not be so useful in solar farms, but e.g. on high-altitude "dwell" drones that don't land, their advantages would be decisive. They are already favored for those reasons on satellites, where they are free of weather. A dwell drone's useful life might not be more than a couple of years anyway, if only because it gets obsolete, and anyway the motor and batteries wear out. Anyway the material does seem to have desirable properties besides just being cheap to make. So, thank you Lea and Sarah, and best of luck!