February 25, 2010 in Business

Fueling innovation

Clean-energy startup faces high demand for reliable generation
Dana Hull San Jose Mercury News
 
Associated Press photos photo

K.R. Sridhar, co-founder and CEO of Bloom Energy, poses next to its power servers at eBay offices in San Jose, Calif. Bloom Energy is a secretive Silicon Valley startup that is working on a new way to produce cleaner energy. Associated Press photos
(Full-size photo)(All photos)

SAN JOSE, Calif. – Bloom Energy, a Sunnyvale, Calif., startup that has been working for years on a fuel cell that would allow homes and businesses to generate their own electricity, officially unveiled its so-called Bloom Box at a highly orchestrated media event Wednesday.

Tech journalists joined Calif. Gov. Arnold Schwarzenegger, Bloom cofounder and CEO K.R. Sridhar, venture capitalist John Doerr and former Secretary of State Colin Powell at eBay’s San Jose headquarters to learn how Bloom, which has raised about $400 million from investors, plans to mass-produce its solid oxide fuel cells.

Google, FedEx and Wal-Mart are among the companies beta-testing the technology; several Bloom Boxes are in use on the eBay campus.

EBay started using five Bloom Energy Servers in July. They produce electricity to power space for 2,000 to 3,000 employees and shaved $100,000 off eBay’s power bill, says Amy Skoczlas Cole, director of eBay’s Green Team. EBay uses natural gas in the boxes but will switch to methane gas from an Oklahoma landfill this spring.

But they’re not cheap: The commercial-scale boxes, which look like a big refrigerator, cost $700,000 to $800,000.

Founded as Ion America, the company re-branded itself as Bloom Energy in September 2006 and is the first clean-tech investment in the vast portfolio of venture firm Kleiner Perkins Caulfied & Byers, where Doerr is a partner and where Powell and Sridhar are “strategic limited partners.” But the highly secretive company has kept a tight lid on its operations until this week.

Fuel cells use hydrogen, natural gas, methane or other fuels to produce electricity through an electrochemical process with a fraction of the emissions of a typical power plant. But cost is still a significant barrier.

There are about six main types of fuel cells in various stages of commercial viability, and experts often break the field into “low temperature” fuel cells that rely on hydrogen and “high temperature” solid oxide fuel cells that can use other fuels like methane or natural gas.

Solid oxide fuel cells operate at red-hot temperatures – usually about 1,000 degrees Celsius.

“Because they operate at high temperatures, they can accept other fuels like natural gas and methane, and that’s an enormous advantage,” said Michael Tucker of the Lawrence Berkeley National Laboratory. “The disadvantage is that they can shatter as they are heating or cooling.”

Solid oxide fuel cells are not new: Since 1999, the Department of Energy, researchers and leading companies have been collaborating on efforts to accelerate the commercial readiness of “SOFCs.”

“Solid oxide has always been the holy grail of fuel cells, and there are a lot of companies working on it,” said Mike Brown, vice president of UTC Power in Connecticut, a division of United Technologies and a leading fuel-cell maker. “The issue is durability and cost. Fuel cells have to be as durable as utilities and operate for at least 10 years.”

Bloom’s secret technology apparently lies in the proprietary green ink that acts as the anode and the black ink that acts as the cathode. Bloom executives would not disclose the composition of the ink. Small cells are then stacked to make a larger device. A key question for Bloom is “stack life”: How long are the stacks going to last? What warranties are they providing their customers?

“What has to be proven by any fuel cell manufacturer is that their technology can operate reliably for years, ideally 10 years, with the ‘four nines’ – 99.99 percent reliability, or very little outages,” said Scott Samuelsen, director of the National Fuel Cell Research Center at the University of California-Irvine and a professor of mechanical, aerospace and environmental engineering. “At this point, Bloom has excellent potential, but they have yet to demonstrate that they’ve met the bars of reliability.”


Thoughts and opinions on this story? Click here to comment >>

Get stories like this in a free daily email