The Powerhouse: Inside the Invention of a Battery to Save the World

by Steve LeVine

  Then there was the proposal itself. As the weeks had gone on, Gallagher’s techno-economic model kept moving closer to the front of the document and now was a growing factor in Argonne’s orals presentation. Just the day before, the group was rehearsing at the lab in front of Isaacs and his deputy Peter Littlewood when Chamberlain reached the middle of the slide deck. The two Bell veterans leapt at Chamberlain’s description of the Nelson-Gallagher model.

  “This has to be right at the beginning,” one said. “This is what we have to talk about.”

  With each such elevation, Gallagher obtained an effective promotion. As it now stood, if Argonne won, he would be catapulted to a management slot just under the senior men.

  Gallagher himself considered the model an “incredibly useful tool.” But he had suppressed his opinion in early Hub discussions, conditioned by the gibes he usually heard around the lab, scorn that the model was “second-rate stuff that anybody who has a bachelor’s degree could do.” The other battery guys shunned the model, which was why Gallagher was “amazed” now as Isaacs and Littlewood talked it up as an ingenious way to build prototypes on the cheap: you would accelerate innovation by quickly informing inventors what would and would not work at factory scale.

  Littlewood’s own excitement went back again to Bell, where he recalled an effort to build a prototype that had inadvertently led to the discovery of the “fractional quantum Hall effect,” a principle of physics that won another Nobel Prize. Littlewood endorsed the Hub’s intention to emphasize prototyping as a first principle.

  To Gallagher, senior management’s attitude clarified the Hub’s core mission, which was “to throw out the old paradigm and say, ‘If you could start from scratch, how would you organize things?’” The writing team was unified in this arguably radical conclusion—that to rapidly launch the age of electric cars, you needed an entirely new system of research. You had to ignore virtually all the battery work currently under way and conceive both a new way to invent and new concepts of storage, a frame of mind that Goodenough himself would recognize. “Most of us believe that you would eventually get there with today’s model,” Gallagher said. “But it would be very inefficient. We want to try to do it in the near future instead of the distant future.” If something out there was going to inadvertently create the big leap, it would have already done so. The only way to win the race was to start over.

  The assertion that nothing currently on the market or in battery labs anywhere in the world was up to the task was a bold one. Gallagher wasn’t even certain that the Hub was the prescription. Unlike Khal Amine, who said that the right team and the right sum of money would produce the battery breakthrough in a decade, Gallagher was “a little more pragmatic. I don’t know. The thing is I am really excited about trying.”

  The Hub might get there, or might not. Gallagher said that battery development as currently practiced at Argonne and everywhere else was a dead end.

  The group had rented and catered an extra room at the Germantown, Maryland, Marriott, where they were staying, thirty miles northwest of Washington, D.C. The orals would be held nearby on Wednesday in the Department of Energy’s basic sciences division. The team would rehearse in the room today—a Monday—then rest tomorrow.

  The team assembled in the morning for this last round of practice. Crabtree went first. The Argonne-led group, he said, had carefully considered the FOA—the announcement of the Hub opportunity—and looked forward to the reviewers’ scrutiny. He flashed a chart on the screen. It was a “unifying graphic” conceived by Gallagher that depicted innovation leading from the laboratory to the market. At the nucleus was the techno-economic model. In the Hub, it was relabeled “system analysis,” and Crabtree, with the fresh orders from the top, promoted it like a breakthrough innovation in itself. It was “a new paradigm for battery development,” he said, an example of “science-based rational design” through which “we can predict what we will get.”

  Crabtree said the current methodology, where scientists published articles and engineers focused on performance, resulted in isolation between groups working on the anode, cathode, and electrolyte. “They operate on a hunt-and-try approach. So if something works well, ‘Let’s do it.’ If it doesn’t work, we will drop it without trying to understand why,” he said. The result was battery performance improvement of about 5 percent per year, “which is actually quite impressive, but not fast enough to achieve the goals.”

  To meet the Argonne team’s transformational goals, there had to be a new way of doing things, Crabtree said. He unveiled five-five-five—five times greater energy density at one fifth the cost, carried out in five years. “So let’s look at how that system would work,” he said.

  As a given battery design was studied, researchers would reach a point at which the Hub managers would step in and caucus: did the results merit a go-ahead? If the answer was yes, the next stage would involve a “translational development team.” Crabtree called attention to a slide. “These colors represent the representatives of every part of the Energy Storage Hub,” he said. This team would now move the promising battery concept along to “more decision points.” If the concept continued to meet the benchmarks, it could then graduate to industrial scale-up.

  The team was not starting with a blank slate of new battery theories. The proposal had three potentially transformational concepts, each of which, in the opinion of the Argonne team, stood a reasonable chance of working. To plumb these concepts, the team had marshaled “game-changing tools,” Crabtree said. They included “the Electrochemical Discovery Lab,” to be equipped with every major measurement tool required to evaluate a battery. These tools were sensitive—you had to be trained in how to use them. But once you were, the tools had the power—along with techno-economic modeling—to carve months and years off the time required to separate winning from losing chemistries.

  Some concepts would not make it. The team would determine that they were unsuitable for development. “In fact, we expect to have many of these,” Crabtree said. Some of the rejects would be nonetheless mined for useful knowledge—lessons learned that would be driven back into the interrogatory process through a specially created “Science Library of Knowledge.”

  “We’ve assembled a team consisting of five national laboratories, five universities, and five industry partners,” Crabtree said. “One team operating under one management with one mission.”

  Chamberlain presented next. He spoke of the “temperature plot,” which Argonne believed would be crucial to the Hub’s success. It was part of techno-economic modeling. The idea was to track every material under study on a chart. Color and line length would indicate the material’s readiness for translation into a prototype. The longer and greener the line, the readier it was in terms of performance, cost, energy, power, safety, and life. Importantly, the data were comprehensive. By cost, Chamberlain did not mean just electrode preparation, coating, and so on, but shipping and receiving the product—the total cost of developing a battery pack. A high-voltage system, for instance, could almost certainly operate in relatively small factories, have a cost advantage, and thus be evident in a long, green line.

  The reviewers needed to contemplate another important question as they compared Argonne with its Hub competitors, Chamberlain said. That was “how quickly can you get going on day one?” Not only did the Argonne team have “the skills and the people set to do it. We already have the equipment,” he said. All the members had signed a legally binding agreement governing intellectual property. “We don’t have this fear” of losing out if an idea is swept up by another team member. Chamberlain said, “We can openly collaborate because we already have a plan.” A single entity—Argonne—would control all licensing. The Hub would be “a one-stop shop for industry,” he said, and “hit the ground running.”

  • • •

  On Wednesday morning, July 18, the Argonne team filed into a battered room in the DOE’s ba
sic sciences division headquarters. Inside sat the reviewers, empty seats for the Argonne guys, and an old projector to display slides.

  The projector didn’t work. Chamberlain summoned Gallagher, who, standing outside the room, had a spare. Then the room went dark—somehow the electricity had gone down. The battery guys stood still. Long minutes passed.

  Finally, the lights returned.

  Chamberlain said, “I think that was eight minutes. We’ll get that back, right?”

  You can have five, a reviewer said.

  The Argonne team silently cropped out lines.

  After the orals were over, the group gathered for coffee. The reviewers, particularly Jeff Dahn, the competing inventor of NMC, had seized on the five-five-five claim. “There is no way you can do this,” Dahn said. Other reviewers picked up the theme. Chamberlain had faced precisely the same doubts at Argonne; before an uprising could break out among the reviewers, he stood and said, “I learned in industry and academia that if you set goals with people or for people, what people do is they naturally try to achieve that goal. So we set it.” He nodded to the reviewers. “It is going to be extraordinarily difficult to achieve this goal. And we understand it is aspirational. But if we make it halfway or two thirds the way there, we will shift the market. And I don’t mean it will move things a little bit and create businesses and jobs. It will shift the market.”

  Chamberlain hoped his point was understood—that if they were ultrarealistic and set a goal for a 50 percent improvement over lithium-ion, the researchers would aim precisely for 50 percent. But if they targeted a fivefold improvement and got twofold, “that is a monumental achievement.” A couple of the battery guys were unhappy with Dahn, but Crabtree thought his questions were fair—tough, but fair.

  The flight back to Chicago was in a couple of hours. At the airport, Isaacs said that no one could have commanded the floor as Crabtree had. When he entered the room, every member of the review committee knew who he was. “We couldn’t have done it without you,” Isaacs said. Chamberlain said that everyone “really did a good job.”

  Someone asked, “Was it a slam dunk?”

  “No, not at all,” Isaacs said. His voice was tight.

  “We are going to wait and see,” Crabtree said.

  It was pointed out that prior to the orals, Thackeray had forecast an Argonne victory. Chamberlain laughed.

  “It is not slam-dunk,” Isaacs said. “There will be other strong competitors and you can’t tell. These committees can do anything that they want to do. And DOE has its own mind. It is going to be a real tough decision.” No one in particular could know what mastery Oak Ridge would put on display.

  Not a slam dunk. But they felt damn confident.

  40

  The Waiting

  Three weeks after the orals, a curt message circulated among Argonne’s senior managers including Crabtree and Chamberlain: assemble for a conference call with Isaacs.

  The call started an hour later.

  “Are you sitting down?” Isaacs asked. “Well, here is a piece of news. We won the Hub.” The long struggle was over. The Hub was Argonne’s.

  Isaacs barely let the news sink in. They had to transform the proposal into working plans for execution as soon as the Department of Energy made the decision public. That could be any day. They needed to be ready. Meanwhile, they were sworn to secrecy. No one else, even other members of the proposal team, could know. Isaacs himself would pass word according to protocol—after the Department of Energy announcement.

  The rest of the team was anxious. Amine said the decision was certain to come in a few days or surely early the following week.

  Amine meanwhile was wheeling and dealing. His immediate goal was to get NMC 2.0 into the iPhone and other Apple electronics. He had met with the company’s battery guys in Cupertino, California. Apple products relied on workhorse lithium-cobalt-oxide batteries. They were unquestionably reliable. Still, they were not the best that Apple could do, he said. Their capacity was 140 milliampere-hours per gram, of which you could access only half because of deterioration over time. The average iPhone could run several hours while using just the telephone but much fewer if you were running Google maps. NMC 2.0, conversely, gave you 190 milliampere-hours per gram, more than 30 percent better capacity. It operated at higher voltage, which meant even greater energy. NMC 2.0 still required work, including coating and additives to protect the surface of the electrodes. But Amine said switching would be a smart strategic move for Apple. The issue was not only performance. So many electric vehicles were going on sale that they would eventually place a serious strain on the global supply of cobalt. The price of the metal was likely to soar. That was not a huge matter for the car-making industry because cobalt made up just 30 percent of the cathode in car batteries in favor at the moment. But the current Apple cathode was 100 percent cobalt oxide.

  Apple dispatched a team to Argonne to examine Amine’s suggestion more deeply. “This is a big deal if Apple licenses. It is huge,” Amine said. “They told us they will need one billion batteries within the next five years. Amazing.” He laughed.

  Two months passed. No Hub decision was announced. The senior team began to wonder whether someone’s mind had changed. Even those in the know said they could no longer be certain of what Isaacs had confided. Maybe they had not won after all. Or perhaps the decision had been overturned. Who knew, given Washington politics and the presidential race?

  “What time is it in Washington now—two forty-five?” Mark Peters asked, standing outdoors one day with Thackeray. “It won’t happen today. It’ll be next week.”

  As the weeks wore on without an announcement, apprehension deepened. Despite assurance from up high, the core group worried that somehow another team had managed in the silence to reverse the decision. All they could do was wait. “If we don’t get the Hub, it will be a major disaster,” Amine said one day.

  In October, A123 declared bankruptcy. It had been on the verge of selling an 80 percent stake to a Chinese company—Wanxiang—when it decided instead to file for Chapter 11 protection. The Argonne team fretted that the bankruptcy could be a political dealbreaker—A123 had received $135 million in federal stimulus money to fund a Michigan battery factory. On the very day it filed the bankruptcy papers, it received a check for $946,830 from the Department of Energy. The Argonne team acted forcefully and told the Department of Energy that A123 had withdrawn from the Hub team. But no one could know whether that would be regarded as sufficient inoculation against political attack should an Argonne victory be announced.

  Crabtree heard from someone in the Department of Energy whom he would not identify that the announcement would probably wait until after the presidential election on November 6. It was easy to infer that no one wanted to risk a scandal that might jeopardize Obama’s reelection.

  Steven Chu was due in Chicago in a couple of weeks. Gallagher hoped that someone would muster the courage to ask who won. Mostly though, Gallagher said he felt naïve. He and the rest of the team had nourished grand expectations for the proposal. They had assumed that outsiders would buy into their idea. Political reality appeared to be stepping in the way. They had never contemplated that risk. “I think it is referred to as ‘drinking the Kool-Aid,’” Gallagher said.

  Obama won easily. Yet there was still no word. “We are going to hear something before Thanksgiving,” Chamberlain said. But there was really no knowing. The battery community was flowing with rumors. Chamberlain said that if he plotted the various purported announcement dates on a graph, the distribution would probably be clotted around the current week—in the middle of November. But some said the announcement would not come for another four months. “I just don’t know what to believe,” he said.

  At last, official word arrived. A conference call was arranged with Argonne management—the lab had won. The announcement would be made November 30 at the University of Chica
go.

  That morning, Chamberlain and Crabtree pulled battery components one by one from bags they toted into the lobby of the Gleacher Center, the modern main building of the University of Chicago Booth School of Business. They were props to explain the Hub’s work to the invited dignitaries. Isaacs waited farther inside, in a staging area for the speakers. Chu arrived before long, eager and alert. Chicago mayor Rahm Emanuel was next. Their bonhomie, forged in Obama administration cabinet meetings, was instantly apparent. Isaacs tried to break in by telling Emanuel he was delighted to deliver the Hub for the city.

  “You didn’t deliver it,” Emanuel shot back. “I did.”

  Emanuel did not appear to be joking. He seemed certain that his White House connections and not the Argonne proposal tipped the decision. Who knew—perhaps Emanuel was crucial in the choice. It did not matter at this point.

  When the ceremony was over, Peters, Isaacs’s deputy, pulled Chamberlain aside. “This was the best moment of my career,” he said. He paused for a long moment, then pointed a finger at Chamberlain. “And it is because of you. Thank you.” Chamberlain felt terrific. He said, “Right back at you, man.”

  Isaacs said an albatross had been lifted. If Argonne had lost, “it would have been everyone’s neck and guess whose fault it would have been?” he said laughing. “There were times when I sweated a lot, but this is what this is—a competition.” Isaacs would make sure that whatever lessons Chamberlain had learned would be codified so Argonne could win future big competitions with less effort—and angst.