In a tidy white lab on the southern edge of Berkeley, scientists are trying to duplicate one of nature's greatest tricks, pulling energy out of thin air.

They're designing artificial leaves that can convert sunlight, carbon dioxide and water into chemical fuel, much like the photosynthesis of flowers and trees.

The team has already built a crude prototype from silicon, polymers and platinum that can create a simple and clean hydrogen fuel. If the scientists figure out how to cheaply produce more complicated energy sources, it would enable mass production of "drop-in" fuels that could power automobiles, trucks, planes and ships without pumping more greenhouse gas into the atmosphere.

In other words, it could provide a viable alternative to digging up more petroleum, coal and other traditional energy sources widely blamed for global warming.

"We have no other option than getting off fossil fuels," said Heinz Frei, acting director of the lab, the north site of the Joint Center for Artificial Photosynthesis. "The research into artificial photosynthesis provides society with an option."

That simple, that hard

Without drastic changes to global energy systems, studies show that rising fossil fuel emissions could push global temperatures up as much as 3 degrees Celsius by 2050 and 6 degrees Celsius by 2100, unleashing a series of dangerous ecological consequences.

Researchers are investigating an array of possibilities for preventing or offsetting certain effects of a warming world, from sucking carbon out of the atmosphere to increasing the reflectivity of clouds. But even those exploring such options say the only way to address the full scale of global warming is to attack the root cause: cutting greenhouse gas emissions as much and as quickly as possible.

"The energy sources we use can't be fossil fuels," said Jane Long, former associate director for energy and environment at Lawrence Livermore National Laboratory. "It's just that simple and just that hard."

Venture capitalists, startups, corporations, government researchers and academic labs around the region are working aggressively to develop or deploy more efficient renewable energy sources, including next-generation batteries, electric cars, light bulbs, biofuels, hydroelectricity, and geothermal and wind power.

Tapping sunlight

But among clean energy options, tapping into the direct power of sunlight is uniquely promising. The sun casts more energy on the globe in one hour than all of humankind consumes in a year. The problem is that existing technology doesn't efficiently capture and store those diffuse and intermittent beams.

That's where artificial photosynthesis comes in.

It could retain more of the sun's energy than biofuels, which for a variety of reasons store well below 1 percent of the energy in the sunlight that touches them. And unlike solar panels, the energy produced by artificial photosynthesis would be highly portable, packed into a dense fuel that could power vehicles over long distances.

"There are lots of ways to provide electricity to things that don't need to take their energy source with them," said Graham Fleming, vice chancellor for research at UC Berkeley and a chemistry professor who has done groundbreaking research in photosynthesis. "But planes, ships, cars and trucks are a different problem and one we don't have a good solution for."

And it's a big problem. Transportation contributes more than a quarter of total U.S. greenhouse gas emissions, and accounted for nearly half of the net increase during the past two decades, according to the Environmental Protection Agency.

The real challenge

During a tour of the Berkeley lab last month, Frei, a slight Swiss chemist with a head of gray hair, stopped at a work space and picked up the closest thing his team has to an artificial leaf.

The "engineering model" looks nothing like its natural counterpart. Four tilted purple strips fill the translucent box, which fits in the palm of his hand.