Biofuel makers struggle, utilities wage war against rooftop solar power and fuels knit together by microbes using carbon dioxide and electricity remain firmly in the lab—but at least there are no oil well blowouts to staunch anymore. Chemist Ellen Williams joined the U.S. Department of Energy’s Advanced Research Projects Agency–Energy as its new head in December 2014. And although some energy innovations currently falter that is nothing compared with the events of 2010 when she joined BP, one of world’s largest energy companies, as chief scientist. That was the year of the blowout of the Macondo well in the Gulf of Mexico that resulted in the Deepwater Horizon disaster, the worst oil spill in U.S. history.
"When I was recruited to join BP as chief scientist, the plum they offered was the opportunity to run their energy sustainability challenge," she recalled at the sixth annual ARPA–E summit on February 10, noting the concerns about the interaction between energy, climate as well as the availability of water, land and minerals. "Energy with climate is the really big connection we have to pay attention to."
As BP notes in its annual analysis of global energy, the human enterprise burns ever more fossil fuels—hence the need to explore for oil in the deep waters of the gulf. The energy company's latest set of statistics foresees no end to that trend and, therefore, continuing increases in the carbon dioxide emissions, among other pollution, that results from all that burning. Rising levels of atmospheric CO2, which have already touched 400 parts per million, trap ever more heat, changing the climate.
Slowing and reversing that trend while ensuring the vitality of the U.S. economy is what ARPA–E was created to do back in 2009, an agency Williams also called "the coolest thing on Earth so how could I possibly stay away?" She added "I'm a true geek. There is nothing I like more than talking technology." Scientific American sat down with Williams at the summit to chat about her vision for the future of energy innovation.
[An edited transcript of the interview follows.]
What are your plans for ARPA–E?
The first thing that I'm hoping to do in this new role—and the most important thing—is to expand our impact. The ARPA–E model of how we do business is unusual, and it's actually unique where we combine cutting-edge technology innovation with a cold-eyed focus on market reality. We bring those two things together, and that's what makes ARPA–E different.
We have developed a pipeline of products in the five years that ARPA–E has been funding projects. Truly innovative things that you never thought were possible have been put in a position where we've shown they can fit into a competitive space. So, moving forward, we're going to see pushing more and more of those products out, developing their impact, understanding their impact and building on what we've learned to design our next round.
Are there particular programs you would like to see? You have an open call right now.
ARPA–E is all about change. It reinvents itself every two to three years. We love our programs and, as I said, they fill up our pipeline. But then they're done. And at the end of the program we're constantly reevaluating and planning the next place to go, so we're going to constantly change and constantly build on our learnings from the past.
Has the ambition changed? Some of the early programs like electrofuels were very audacious whereas some of the more recent programs like methane monitoring are perhaps less so.
I think we've got a great balance and a great opportunity to sustain that balance. So electrofuels was, as you said, very early-stage, but it had the benefit of nucleating an entirely new field. Our construct is really to drive forward technology and evaluate what it would take to bring it to market. Electrofuels was a success because we now understand what it's going to take, and it will take awhile for electrofuels to reach commercial readiness.
So that would be one end of the spectrum. And then we move all way towards programs where there's a real new advance that's going to make it possible to a much more efficient engine. And that's an equally valid part of ARPA–E's portfolio.
Are there particular past successes or new projects that helped attract you to ARPA–E?
So you know we have an aging grid. I mean people say that Thomas Edison would recognize this grid. There's nothing in this grid that would surprise Edison in how it works and how it's set up.
In our lifetime, however, everything else in our lives has been completely revolutionized by the digital revolution. Now the grid is certainly using more digital technologies but there's a huge opportunity there. It's such a big, complicated system and it's facing so many more stresses in terms of extreme weather, changing usage patterns and integration of distributed and renewable generation.
ARPA–E by itself is not going to solve the grid. We don't do road map problems. What we do do is look at everything that's already going on out there and look for an opportunity in the space that no one else is looking at.
So we're looking beyond demand response activity and saying: "Can we really look at the grid as a dynamic system where we tie together all demand, all generation to really tailor the demand and response to work in synchrony in an effective way?" So this very exciting opportunity brings together applied mathematicians, computer scientists and experts in the power grid in a way that hasn't been done before. And we're going to give it a push, ARPA–E's specialty, giving a hard push in an area of high-risk opportunity and see if we can break that loose. If it goes, it will have a big impact on the grid overall.
Do you have enough money to take those high risks? I know from talking to past directors, like Arun Majumdar, that he was always wistful that he couldn't do anything in nuclear because he just didn't have enough funds.
Was he wistful?
Well, that's what he expressed: "That is something that I wish I had had the budget to try."
Interesting. Well, we've got a program structure around this idea of giving a sharp push into a well-focused area of opportunity. So we are well structured to use the budget we have very effectively, because we can tune the number of projects that we undertake to match our budget. More is always better, but we can do a great job with what we have.
Coming in from the outside, are there particular failures that ARPA–E hasn't learned enough from?
We absolutely kill projects. I would say it's rare that we've killed a project because it was actually a failure. We're trying high-risk projects because they're high risk. And we need to de-risk them. If we discover going forward through a project that this one approach isn't working…. I mean, that in itself is not a failure—that in itself is learning. It would be a failure if we kept pushing, trying to get to a goal that we couldn't get. So each of those times that we've canceled a project, in general, it's meant that we've really learned something.
Are programs in biofuels and the like at risk given the current low cost of oil?
We have to keep in mind that ARPA–E is about options, and oil prices are low right now but you've lived long enough to see a few oil shocks—I've certainly lived even longer, I've seen quite a few oil shocks. So nothing's certain but I'll say that I'm as certain as these things come that there will be another oil shock. We're about options at ARPA–E and making sure those options for the future are on the table so we're not caught flat-footed.
Are there options like solar that ARPA–E no longer needs to work on, thanks to recent price drops?
If you come back to our model of working, we'll look at the big energy system and we're always looking for a place that has an opportunity or a need. And if there's not an opportunity or a need, we're not there. So we'll stop working in that area.
We have solar projects because we see a different way of addressing the problem, and it has the potential to have a big impact. Again, more options for the future.
Who is the customer for these options?
Much of our structure draws from [the Defense Advanced Research Projects Agency], and DARPA has a customer—DARPA has the Department of Defense and that is one way in which we are very different from DARPA.
To a large extent our customer is the U.S. economy. If you really read our founding language, you will see that it's very clearly stated that we're here to help the U.S. economy, to improve technical competitiveness, to improve energy security. So I think it's fair to say that our customer is the U.S. economy.
What can you achieve at ARPA–E that you couldn't do a major global oil company or in academia?
As an academic, I was involved in very fundamental research. I started out my career taking my data with a pen that made traces on a piece of paper. I lived the digital revolution and saw firsthand how it's completely changed what we can do and research. So that was a great aspect of being an academic. But, especially being very fundamental with my work, I wasn't going to be able to see the immediate transformation of things that I had produced in a laboratory into the marketplace.
At BP, it was just a great opportunity to learn about the energy system. It was a stupendous opportunity for me to see what it takes in a corporate environment to move a technology from concept into implementation. Especially in a big company like BP, the requirement on impact was pretty big. You know, a small project that was going to make a small amount of money was probably not something that was going to be a major focus.
What we have here at ARPA–E is, I think, the best of both worlds. From the point of view of someone who loves innovation and loves seeing lots of new technologies, we have the ability to try a lot of things that are high risk, that are very innovative and identify the ones that are going to go forward and have big impacts. So we have both impact and the ability to look at lots of different options.