ARPA-E Energy Transformation

Some 1,700 energy aficionados from 15 countries and 49 U.S. states filled a huge convention hall in suburban Washington, D.C., in early March for the first public meeting of the Department of Energy’s Advanced Research Projects Agency-Energy (ARPA-E).

Attendees included an enthusiastic mix of scientists, engineers, venture capitalists, technology developers, energy entrepreneurs, and state and local government officials. They came primarily from industry (59%), but also from universities (21%) and government (17%). More than 40% hold Ph.D.s. All had the goal of trying to understand how this new and unusual DOE program works and how they can take advantage of the research dollars ARPA-E has to offer.

To wit, a workshop on the first day of the three-day conference was billed to provide applicants with “strategic insights” into the ARPA-E program’s direction and its “future funding opportunities,” and it drew some 900 participants. Although top ARPA-E staff did lay out program details, the audience size was far from the intimate exchange some attendees told C&EN they had hoped for.

From ARPA-E’s beginning, DOE has been overwhelmed by interest in this program and its ambitious goal of finding, funding, and overseeing development of high-risk transformational energy-related technologies. The technologies are too risky for private funders but could drive a clean-energy revolution and make the U.S. the global leader in new energy markets (C&EN, Jan. 4, page 19).

Speaking at the conference, ARPA-E Director Arun Majumdar explained his plan to find the best “game changing,” “disruptive” energy technologies, give them funding, and then use his expertise and that of his small staff to develop the technologies and guide them to commercialization. He described an active interaction between his staff and the technology developers that is unique for a government program.

Majumdar has been in his position for only four months, and ARPA-E itself is new. But it has moved fast for a government agency, and at the conference Majumdar announced a third round of funding solicitations. With that solicitation, ARPA-E has allocated $350 million since last April, when it was first given funds to distribute.

Although Congress created the program in 2007, it never supported it. Funding fell to Energy Secretary Steven Chu, a Nobel Prize-winning physicist and ARPA-E cheerleader, and President Barack Obama, who set aside $400 million from the American Recovery & Reinvestment Act of 2009.

Shane Kosinski, ARPA-E deputy director for operations, was one of the first DOE staff to come on board in April 2009. He told conference attendees about DOE’s preparation for its first solicitation that took place that month when it sought applicants for a pool of $150 million.

DOE was fishing, he explained, and offered a general solicitation. “Our intention was to use the results of the first solicitation to set the direction of what we should focus on in the future,” Kosinski said.

The department expected about 500 to 600 project applications, Kosinski said, but he had been warned that ARPA-E might receive as many as 1,000. That amount, he said, was their worst-case scenario and was likely to break the system if it happened.

When applications came due in June, ARPA-E was swamped with 3,700 concept papers and began a fast search for what ended up being some 500 reviewers, Kosinski said. Staff whittled down the 3,700 hopefuls to 350 or so applicants who were asked for full proposals, and in late October the agency announced 37 finalists—1% of applicants. ARPA-E staff members have negotiated contracts with all but two of those winners for grants averaging $4 million to be spread over three or four years.

To help guide the program since that first solicitation, Kosinski said, ARPA-E has held seven invitation-only, technology-specific workshops, limited to fewer than 100 people. The March conference was the first event open to the general public.

Majumdar noted that in December DOE announced a second round of solicitations for another $100 million. This round was much more specific and was directed to three program areas: liquid fuels, carbon capture from coal-fired power plants, and batteries for electric vehicles (C&EN, Dec. 14, 2009, page 28).

As a result, this time ARPA-E received a more manageable 500 concept papers. The full applications were due on March 15, and final selection is expected in early spring. DOE expects between 10 and 15 winners.

For the third round of grants, Majumdar noted, the pool is $100 million. In this round, DOE is specifically seeking concept papers for grid-scale electricity storage technologies, advanced power converters, and new advanced energy-efficient cooling technologies, particularly those that can be added to existing cooling systems in the developing world. Applicants are to file five-page concept papers to DOE by April 2 (C&EN, March 15, page 40).

Majumdar and ARPA-E staff explained that the review process is complex and interactive. And for winners, the review continues throughout the project funding cycle, which could run as long as four years.

Once selected, each project is assigned a dedicated ARPA-E team made up of technical experts as well as lawyers, venture capitalists, manufacturers, and others to help move the project to commercialization. The ARPA-E team and the project winners negotiate a set of milestones to guide the projects over the years of funding. Quarterly site visits by ARPA-E staff are required, and annual “go, no-go” decisions are part of the process.

Majumdar played down the risk of failure after receiving funds, saying it should be expected with such high-risk projects. ARPA-E’s goal is not “incremental change,” he stressed, but a transformation. He added that ARPA-E staff would help find new applications for technologies that are rejected after obtaining funds.

Along with Majumdar and Kosinski, the three program managers also spoke at the conference workshop. The program managers are key to ARPA-E’s success because they, along with Majumdar, select the winners and lead hands-on interactions with successful project developers. How this small staff is supposed to accomplish all these goals was not discussed.

However, when the five took questions, lines 10 people deep quickly appeared at the microphones. Inquiries focused on a perceived need for more funding, how success will be measured, the paperwork load that may come with quarterly assessments, and the lousy odds of being selected for funding in the first place.

Not all questions were answered, however. On funding, Majumdar stressed that it was up to Congress and the President.

Majumdar acknowledged the difficulty in measuring success in a program in which transformational projects may not show a payoff for 15 years. But he said success would be measured by whether projects obtained private funding after ARPA-E money ran out and if ancillary technologies were developed to complement and support the original ARPA-E-funded technology.

Paperwork, the program managers stressed during the question-and-answer session, would be minimized once projects are selected. They also said concept papers are to be five pages long and full applications are limited to 40 pages.

Several speakers stressed the 1% odds of winning a contract. Majumdar responded that the low level was for the first round and odds will get better as ARPA-E more clearly defines its goals. He added, however, “About 20 to 25% of the first applicants presented technologies that defied the first two laws of thermodynamics.” But still, “about 2,000 were valid proposals,” he said.

A representative of the venture capital investment firm Kleiner Perkins Caufield & Byers said that in his experience a 1% success rate for this kind of high-risk program is “about right,” and lately even that level has been declining.

Talking outside the hall, many participants said they applied in the first solicitation round and failed to receive a project grant. They complained of being unsure why their concept papers fell short and said DOE provided too little information on rejections to allow them to improve their next application. “Instead of telling us what we ought to do,” one attendee told C&EN, “I’d like to know exactly why the program managers rejected a project.”

Just off the conference floor in a barnlike space, some 130 booths showcased energy-related companies and technologies. Most were small start-ups with unfamiliar names, but among them was Dow Chemical, one of the largest and oldest U.S. chemical companies.

“Branding” had much to do with why Dow was there, explained George R. Hamilton, vice president of government business development. “We see ourselves as one of the largest industrial users of energy and a leader in advanced materials. We are motivated and capable of solving energy challenges.

“We want people to think of Dow as more than a chemical company and as an energy solutions provider,” he continued, pointing out several chemical-related products Dow is developing and selling, from solar shingles to adhesives to lightweight chemical composites for automobile manufacturers.

Dow has also unsuccessfully applied for ARPA-E grants, and several technology developers with pockets that are not as deep as Dow’s have questioned why companies its size should be involved in ARPA-E.

To that, Hamilton said: “Private companies have investors, and their money goes to where there is a good return with a reasonable level of risk. For some of these areas in alternative energy, without a price on carbon emissions, the market just isn’t there yet. A multi-million-dollar investment won’t pass the hurdle for Dow; the risk is too high.”

The small ARPA-E investments, he says, help keep the U.S. in the game while the market develops. “The government has a role to play in all this. The Administration is doing the right thing.”