Solar-powered soldiers in the pipeline

Nine years ago, Robert Nowak, an electro-chemicals expert for the US Defence Department, learned that senior generals weren't happy with their troops' electronic gear. While the night vision, laser, and GPS devices worked well, the batteries that powered them weighed some 25lb per soldier, heavy enough to hurt some of the troops.

So Nowak, who worked at the Defense Advanced Research Projects Agency (Darpa), the Defence Department's research branch, solicited bids for a new device that would power a soldier's gear at a 10th of the weight and a fraction of the $100 (£60) cost of the batteries. Today, the original 18 companies that took up Nowak's challenge have been whittled down to two: Livermore, California-based UltraCell and Adaptive Materials of Ann Arbor, Michigan. Their solution: small, sturdy fuel cells that can power a soldier's clutch of mobile devices for a week on a gallon or so of methanol or propane. Battle-ready versions of the fuel cells will be available this year.

Darpa regards the result as a game-changer for the military – akin to the potential shift in the automobile market from petrol-driven to hybrid or electric cars. Before the fuel cells: "If you were in Afghanistan and had a battery, you basically had to go to another country to get it recharged," says Nowak, who is now retired.

Consumers and businesses might someday gain as well. Both companies are testing models for the US commercial market. First targets: city police forces and makers of recreational vehicles.

The big drive to create a viable alternative-energy future – by Detroit, multinationals such as IBM and BP, and Silicon Valley startups – is well known. But there's another serious player in this sphere: the US military, and especially Darpa.

Created at the height of the Cold War to bolster US military technology after the Soviet Union's Sputnik satellite launch, the agency has a long history of innovation. Most famously, Darpa's researchers first linked together computers at four locations in the early 1960s to form the Arpanet, a computer network for researchers that was the core of what eventually grew into the internet. Other breakthroughs have lead to the commercial development of semiconductors, GPS, and Unix, the widely used computer operating system. There have been some serious gaffes as well, including mechanical elephants to carry troops through Vietnam's jungles and an ill-conceived search for people gifted with psychic powers. But on the whole, Darpa has a strong record of bringing ideas from the lab to the real world.

Can Darpa now score another double success by changing how both the military and civilian worlds consume and produce energy? The Arlington, Virginia-based agency's first goal is always to magnify the might of the US armed forces. That's why it is devoting an estimated $100m of its $3bn annual budget to alternative energy. The US forces deployed in Afghanistan and Iraq are voracious consumers of energy. As a result, they have become perilously dependent on long, costly, and vulnerable convoys of diesel-fuel tankers. More vehicles are used to transport and guard the fuel – mostly for running generators for air conditioning, laptops, and other gear at US bases and posts – than are deployed in actual combat, according to a May report by the Military Advisory Board. With the expense of convoys and guards thrown in, the cost of a gallon of fuel used at the front can range from $15 to several hundred dollars, says the same report. So the army has set an overall goal of significantly reducing its fuel requirements. Under its plan, fuel and supply shipments to 5,000-troop brigades would be reduced from the current once every few days to just one a month.

Darpa describes itself as an incubator of long-shot technologies too risky for almost anyone else to take on. The agency operates by issuing challenges to companies that are so tough they are called " Darpa-hard". Typically, Darpa requires contractors to come up with solutions that are orders of magnitude superior to current technology. It pays companies – from startups to IBM – as well as top universities to meet a goal. Then, other than imposing strict reporting requirements, the agency gets out of the way of the researchers' work.

In addition to spurring the development of palm-size fuel cells, Darpa has contracted with companies to miniaturise solar cells that would supplant the need for generators. It now wants to develop inexpensive diesel and jet fuel from algae that could be produced in the battle zone. All three programmes include the aim of accelerating the manufacture of any new product by private companies, from whom the military could buy.

The agency certainly has no shortage of ambition. Take its solar panel programme. Current technology converts into electricity just about 20 per cent of the sunlight that hits silicon panels. DuPont and the University of Delaware are partners in a Darpa contract worth up to $100m to elevate efficiency to 40 per cent, at an affordable price. The idea is to capture the sunlight that would normally fall across a square-metre solar panel and concentrate it into a cell about the size of a fingernail. A number of those miniaturised cells would be arrayed across a panel that could be folded up and toted into battle, where it would power the needs of a half-dozen to a dozen soldiers. "It's an aggressive goal," says Brian Pierce, who is managing the Darpa programme. In contrast, solar cell maker SolFocus of Mountain View, California, is working on similar technology for civilian applications but is aiming for much more modest efficiency gains.

Darpa wants the cost of the new panel not to exceed $1,500 – compared with the more than $15,000 DuPont recently spent on a working model of the panel and its cells. Dan Laubacher, DuPont's manager for the project, says the system is at least two years away from delivery to the military. But as production ramps up, he says, the ultimate cost "could be lower" than the $1,500 targeted by Darpa. Eventually, as costs come down, DuPont hopes to sell the panels to utilities.

Darpa-inspired fuel efforts would change the military. How much the agency could change the commercial alternative energy industry is a matter of debate. Some in Silicon Valley welcome Darpa's commitment. Vinod Khosla, a co-founder of Sun Microsystems and one of the most active venture capitalists in alternative energy in the valley, notes that so far both the private and public sectors have failed to make a definitive breakthrough in alternative energy. "Nobody knows the right answer. So the more the merrier," he says. "Darpa's ability to take a long-term view of research is positive."

However, some argue that alternative energy is unlike other Darpa efforts in the past, when the agency had a tremendous impact. In nurturing a proto-internet, for example, Darpa was alone in the field. Now hundreds of companies are exploring solar panel technology, doing advanced battery research, and experimenting with algae-based biofuels. This is also a global field, where Japan, Germany, and China already have the lead in critical areas.

Others say Darpa's goals can be unrealistic. Darpa wants to reduce the current cost of algae-based fuel from $20-$30 per gallon down to $3. In January, Darpa awarded contracts worth up to $34.8m to two companies to produce aviation fuel at $3 a gallon from algae. The competitors are General Atomics, best known for its Predator drone, and Science Applications International. They have three years to do it. Some doubt these companies – or any company – can achieve that goal.

Chris Somerville, the director of the BP-funded Energy Biosciences Institute at the University of California Berkeley, has specifically avoided investment in algae-based fuel because his team does not see costs dropping below $10 a gallon. "We're sceptical that that's going to be possible," Somerville says of the $3 price target. Darpa's answer, as expressed by Nowak, is simple: "If you want to change the world," he says, "set the bar high."

This story also appears in the current issue of BusinessWeek