For capturing the sun's copious energy, there are basically two available engineering models: photovoltaic (PV) cells that turn it into flowing electrons or photosynthetic plant cells that turn it into plant food. So which does the job better? After all, such a judgment might help inform policymakers on whether to pursue biofuels or solar electricity.

But the question admits no easy answer, because it begs the deeper question of which one values more: the sheer quantity of electrons produced—so-called efficiency—or the transformation of sunlight into stored chemical energy? After all, storage is a high-value proposition that has made fossil oil, originally derived from plants, so valuable—cheap, energy dense, easy to transport and storable for later use. That is not the case for electricity from the sun—or any other source—which must be captured the instant it is produced and currently has a limited and expensive option for storage: batteries.

"Chemical fuels [hydrocarbons, like those in oil] would be the game-changer if you could directly make them efficiently from sunlight," notes chemist Nathan Lewis, who directs a lab focused on just that prospect: the U.S. Department of Energy's Joint Center for Artificial Photosynthesis. "It pairs the biggest source of energy and the biggest storage."

So, a group of 18 biologists, chemists and physicists set out to answer the question by first creating roughly equivalent systems—comparing apples with apples, as it were rather than apples with oranges. Photosynthesis (conducted by algae) turns roughly 3 percent of incoming sunlight into organic compounds, including yet more plant cells, annually. "Artificial photosynthesis"—comprising a PV cell that provides the electricity to split water into hydrogen and oxygen—turns roughly 10 percent of incoming sunlight into usable hydrogen annually.

That discrepancy suggests there might be room for improvement in photosynthesis, according to the analysis published May 13 in Science. After all, solar cells are capable of absorbing more of the energy in sunlight because they capture it across the electromagnetic spectrum ranging from infrared to ultraviolet, whereas chlorophyll and other photosynthetic pigments absorb only visual light. Introducing pigments to plants that would help them capture ultraviolet or infrared light could change that equation.

Another idea would be to reconfigure photosynthesis itself. Presently plants employ two systems—dubbed photosystem I and photosystem II—to convert sunlight, CO2 and water into carbohydrates. But both of these photosystems rely on capturing visible light photons, which means the two systems compete for each incoming ray of sunlight. If scientists tweaked the system so that photosystem I relied on visible light but II absorbed, say, ultraviolet light—the efficiency of plants would improve considerably. 

"It would be the biological equivalent of a tandem photovoltaic cell," or the stacked photovoltaic cells that absorb different wavelengths of light, says biochemist Robert Blankenship of Washington University in St. Louis, lead author of the analysis. Stacked PV has been demonstrated to convert more than 40 percent of incoming sunlight into electricity, albeit at a prohibitively high price. Such synthetic photosynthetic organism could then become the fuel refinery of the future—a prospect being actively pursued by the Advanced Research Projects Agency–Energy (ARPA–e), a recently formed federal agency tasked with taking scientific findings on alternative energy and turning them into deployable technologies.

At the same time, any biological sunlight-capture method faces one significant constraint—the enhanced bugs or plants have to be kept alive. "We don't want them using those resources to make bugs; we want them to use them to make fuel," explains chemist Eric Toone, ARPA–e's deputy director for technology and program manager for so-called electrofuels—an effort to harness extremophiles to make fuels for human use—who was not involved in this analysis. "As you tinker with bugs to turn off pathways that aren't doing what you want them to do, you've got to leave the bug capable of staying alive."

Nor did the scientists consider other factors that could diminish the utility of either or both approaches, such as land or water needs, waste, impacts on food supply or any of a host of other relevant considerations. For example, the fact that hydrogen fuel-cell cars still cost hundreds of thousands of dollars might overwhelm the usefulness of artificial photosynthesis to produce the lightest element. Still, simply on the basis of converting the most sunlight to usable energy, artificial photosynthesis wins.

But don't count out nature, enhanced or otherwise, yet. After all, plants do several things very well that photovoltaic cells—or artificial photosynthesis systems—do not, such as absorb CO2 at low concentrations (382 parts-per-million and rising) directly from the air and use sunlight to turn it into fuel and oxygen.

"Natural photosynthesis turns CO2 into sugars with lots of carbon-carbon bonds," says chemist Andrew Bocarsly of Princeton University, who was not involved with the analysis. "We've been studying CO2 chemistry for a long time, more than 100 years, and there's very little evidence that we could do what a leaf does."

Of course, plants also have another significant advantage—a bad photosynthetic cell can repair itself; in fact, that's part of its normal operation. No artificial system yet devised—super-efficient or otherwise—can heal itself.