Some 2.4 billion years ago, tiny blue-green algae figured out a neat trick. Using sunshine, water, and carbon dioxide, they produced plant food as well as the oxygen that makes our existence possible. We're still driving around on the fruits of their labor all these eons later-not to mention turning on the lights, flying into space, and everything else we do with the energy embedded in the fossilized sunshine that is coal, oil, and natural gas.But plants aren't all that efficient at harvesting the energy that falls on them as sunlight, so the chemist Daniel Nocera of MIT is trying to mimic photosynthesis, and improve on it. His idea is simple: Split water into hydrogen and oxygen with sunlight, and then recombine them (which creates energy) in a fuel cell when power is needed. The trick is to do both these things cheaply.Nocera has come up with a way to split water into its constituent elements that is less expensive than the machines used today. But he has yet to show that it can be done on a broad scale.For now, the future of an artificial leaf that's as cheap as the real ones may come down to some elementary chemistry: Can scientists find replacements for the expensive and rare metals currently required? Nocera says a "new catalyst" for fuel cells is what's needed, but he's hopeful that he'll find it. He predicts success in less than a decade-or about as long as it takes to grow a tree.