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

by Steve LeVine

  It was about this time when Gallagher was hired to learn Nelson’s system. The pair began to meet with automakers and battery companies in order to obtain more precise estimates. The software’s accuracy was soon impressive. When the pair decided to give it away on the Internet, posting it on the Argonne Web site, they attracted attention from industry players who perceived a new way to cull practical ideas from the less so. Other labs, including MIT, adopted it, too.

  The Nelson-Gallagher model was far from standard procedure at Argonne itself. The senior managers understood its merit, as did their supervisors in Washington. But if you asked Amine whether he had vetted his new ideas with Nelson, he would stare as though you had suggested checking with Santa Claus. Trahey, Gallagher’s own office mate, never inquired as to the feasibility of her concepts. That depressed Gallagher.

  “The group is not interested in seeing how their work fits into the big picture,” he said.

  Trahey said that the other battery guys were simply “overwhelmed” with their various responsibilities. “They care. They don’t want to be working on something futile. They want to know if it is something totally futile,” she said.

  “I don’t think they do,” Gallagher said.

  Gallagher was right. Trahey was merely trying to salve his feelings. The model was sly. To the degree it was actually applied, the software could undermine the proclivity of some battery guys to inflate the potential of their inventions. It amounted to a stern manager peering over their shoulder.

  As Gallagher understood it, the Hub would change all that. It was meant to be a dynamic research environment in which, if an idea was not working, resources would be more or less automatically redirected to something that might. To Gallagher, the Hub was a magical thought: Argonne had to win it. And he had to be in it. Here would be an honest appraisal of the battery work. The battery guys might never attain the desired electric-car performance; they might not win the battery race. But if they could not be straight with themselves, they definitely would not reach the goal, he said.

  Gallagher’s idea, embraced by Chamberlain, was to embed the vetting model into the Hub’s management system. Armed with this data, a battery guy could peer deeper and isolate the specific materials having the best physical properties for the desired performance. The managers began to call the work “techno-economic modeling.” Nelson’s name dropped out of the discussion. Gallagher himself mentioned his former mentor’s role almost only in passing. It wasn’t that he was not grateful. “But at some point you have to hand it off so young folks can run with it,” he said. Nelson did not seem to begrudge the treatment. It’s what it meant to be one of the old people.

  Gallagher was still uncertain about what the model could achieve. It was one matter to write such a system into the Hub proposal. It was another “to actually go and do it.” It would be a new way of business for most of the lab and would require strong leadership to become an accepted tool.

  One thing he knew was that most of the researchers did not believe that vetting was “as big of a deal as it actually is.” They did not understand that if you could determine what did and did not work, that was “a clear stepping-stone” to the better battery. Perhaps the Hub would change their minds. The Hub’s industry partners were certain to press the case. They would be biased toward such testing and intolerant of any inclination to “hide behind things. ’Cause if you make a bad product, everyone knows it,” Gallagher said. “That’s the thing—in the labs, you can hide behind things. You can play games.”

  Gallagher said that Envia’s new battery should be subjected to the model. He had been in touch with Sun-Ho Kang, the former Argonne researcher now at Samsung. Kang’s supervisors were inquiring about the ARPA-E announcement, and he wondered what Gallagher thought of the reported advance. Gallagher said he was skeptical. But he had managed to create philosophical distance from his early rejectionism. “It’s natural that you say two things,” he said. “First, ‘Why couldn’t I get those results?’ Second, ‘It’s great that Envia did’ because you had a role in creating this material that will be in the market.” But one way to know for sure would be to examine the detailed data.

  If the techno-economic model upset some people, “well, they shouldn’t be in the Hub then, right?” Gallagher said. “The Hub is supposed to be a different way of doing business, and if you’re not interested, then you don’t have to be a part of it.”

  Gallagher’s conviction was that Chamberlain and the other senior managers were not the only ones with careers on the line. Here was his chance to spend the next part of his life encouraging scrupulous work.

  • • •

  Chamberlain pulled up a story on his screen. That morning, gas had exploded in GM’s experimental battery lab outside Detroit, blowing out three windows and forcing open a fortified door. Fortunately, no one was killed and just one worker had to be hospitalized. It happened during the extreme testing of an A123 battery under consideration for GM’s future pure electric Chevy Spark.

  “This isn’t good,” Chamberlain said.

  The United States was not alone in its suffering. Notwithstanding the bravado in Orlando, Japan was experiencing angst over the demise of “Panel Bay,” a densely packed strip of electronics factories three hundred miles southwest of Tokyo on Osaka Bay. After years of success, Panel Bay was now seeing Sharp, Panasonic, and other plants either shut down or sold off to Taiwan and other foreign buyers. Sony had lost its way. The edge was now held increasingly by South Korea and China.

  But this did not cheer the battery guys, especially since the bad news was still piling up. Five days later, a forty-eight-page report by the Union of Concerned Scientists cast doubt on the green footprint of electric cars. The report noted that the cleanliness of a particular electric vehicle depended on where it was charged up. If its owner lived in a coal-burning state, the coal used to produce the electricity that charged up the car might be responsible for even greater emissions than a gasoline-fueled model. That is because oil and gasoline generally produce just half the CO2 as coal. If the same electricity was created in a natural gas–burning plant, it would be a different story because it emits only one third of the greenhouse gases as coal. The report said 53 percent of Americans resided in coal-burning states. A New York Times headline asked, “HOW GREEN ARE ELECTRIC CARS? DEPENDS ON WHERE YOU PLUG IN.”1

  35

  Red Team

  The Hub was intended to emulate Bell Labs. But it also had to eclipse Bell’s achievements. It had to both create advances and parlay them into commercial products. Chamberlain said it could not “do science only for the good and noble sake of science.”

  Eric Isaacs, the former Bell manager who ran Argonne, said the labs had their own Achilles heel. They failed to kill theories that weren’t panning out, what industry players called “down-selecting.” As the techno-economic model had highlighted, a battery guy typically conceived of a new electrode, attracted Department of Energy funding, and continued developing it as long as the money held out, whether or not it showed good progress. Consider voltage fade—it took the Battery Department more than a year to organize into a working team after Washington ordered it to. Chamberlain said, “That would never fly in industry.”

  The difference, Chamberlain said, was how government and industry scientists were evaluated: In the Battery Department, you were measured by the number of papers you published, how many times others referenced your work, your awards, and, in rare instances, the transition of your stuff into the market. In industry, the marker was singular—the company’s financial performance.

  GM’s Mark Mathias worried that if the Department of Energy was not careful, the Hub could “end up being a bunch of research papers.” Mathias led the automaker’s electrochemical research lab outside Rochester, New York, and exercised considerable influence over whether GM would participate in the Hub. In order to succeed, he said, the Hub would have to be “run a li
ttle more like a business than is traditionally done in the national lab.” Otherwise there was no point to collaborating with it.

  There was, of course, the converse risk of down-selecting excessively, not providing the chance for promising ideas to bloom. The Hub had to settle somewhere in the middle—it had to cultivate creativity while ultimately culling out what was not working.

  Chamberlain intended to “make swift decisions when we see something succeeding or we see something failing.” That was not his current reputation. He was known to go to great lengths to get along. If he was deputy director of the Hub, which seemed likely, he would at once be thrust into the position of hatchet man—he would have to kill projects. But he said he was ready. “I’ll do it,” he said. Knowing himself, Chamberlain had organized the ground in advance so that he would not appear to be an ogre. He embedded incentives in the Hub structure to encourage down-selecting, or at least lessen the resistance: if your project failed—if you thought it should be killed—you could say so yourself and obtain first dibs on the next big high-risk project. The Hub would thus reward a researcher’s proactive exit from an underperforming concept. Chamberlain would not have to be the one to say no. But it was a fair question as to how long he could maintain that front before he was found out and had to make clear his decisive hand in the down-selection.

  • • •

  Thackeray and his wife, Lisa, were sitting down at home for a dinner of pasta. His cancer had recurred, engorging his spleen with blood and distending his stomach. His doctor had prescribed chemotherapy and he had just undergone his final treatment. He spoke slowly and after dinner stretched out on a La-Z-Boy. Some 40 percent of his blood was concentrated just in the spleen, leaving the rest of his body to survive on the remaining 60 percent. “I feel guilty. I feel I have let everyone down with the Hub,” he said. Isaacs, the Argonne director, had wanted him to run the Hub, but he simply wasn’t up to another intensive five-year job. Isaacs seemed to understand, but Thackeray still often raised the subject.

  Regardless of the result of the chemo, Thackeray’s doctor was recommending that the spleen be removed. He wanted to conduct a biopsy. If Thackeray had the type of cancer that the doctor suspected, the organ was better out. If he did not, there would have been no need to remove it. But all in all, Thackeray said, “they say it is better to have it out.” One could live without a spleen. And the doctor promised that he would “feel like I did five years ago.” A sparkle seemed to appear in his eye. His first plan after recovering was to start on his book.

  Isaacs was thinking ahead regarding the Hub competition. In order to help ensure that Argonne did not blow it this time, he built professional critics into the proposal writing process. These were gadflies practiced at organizing militaristic “Pink Team” and “Red Team” reviews and ripping proposals to shreds in order to get them right.

  Holly Coghill was an expert manager for big-ticket Department of Energy proposals. She was steeped in the national labs—her uncle once ran the Pacific Northwest National Laboratory and Mark Peters, Isaacs’s deputy, was a cousin. She knew the inside personalities like they were family, including what they favored in such proposals and how they preferred them to be organized. A blonde with short-cropped feathery hair, Coghill favored Levi’s jackets and sent out a signal not to trifle with her. One Hub interlocutor, Coghill said, was “all sizzle and no stink. There’s just not a lot of depth there.” When in her view an outsider spoke disrespectfully to Chamberlain, she replied, “Don’t mess with my kids or I’ll freakin’ kill ya.”

  But Isaacs’s main secret weapon was Bill Madia, a nuclear chemist who had run two national labs and established a victory streak in the competition for large government projects. The sixty-four-year-old Madia was an inflated, difficult-to-like egotist with little apparent interest in getting along apart from with those possessing influence over his next project. But he also was “the godfather of the national labs,” Coghill said. He was ingenious at winning big-money, prestige-building competitions, making people forgive his foibles. This promised land of scientific shrine making was so coveted among the research elite that those able to lead the campaigns earned enormous fees and permission to exhibit almost any pathology. In all, the gadfly budget for Argonne’s Hub proposal was about $500,000.

  Isaacs was relying on Madia. “He has won multiple times,” Isaacs said. “He just gets it.” Isaacs wanted to win. He wanted to rid Argonne of its long, unlucky history. Madia was going to get him there.

  He knew Madia’s habit of making others miserable had already triggered an uprising on Chamberlain’s team. They wanted him out. But Isaacs was unmoved. Madia was indispensable. If he was tough, Isaacs wanted him to be. “The most important thing,” he said, “was (to make) sure that Jeff was on his toes.”

  • • •

  The Pink Team’s duties were clear. Its members were to appraise the draft proposal and provide recommendations on how to improve it, as a preliminary to a later, more exacting “Red Team.” They also were to “score” it. Department of Energy judges evaluated proposals according to criteria set out in the Funding Opportunity Announcement, the contest declaration known as a FOA. The criteria—minutely detailed under broad categories such as scientific and technical merit, competency of the team, and reasonableness of the budget—were scored on a scale of one to one thousand. The Pink Team would provide Chamberlain with an assessment of Argonne’s current standing.

  The assessment lasted two days. Madia was harsh. Argonne’s vision—its “story”—did not shine through. The narrative was “buried deep in the science.” The scientific sections were adequate as far as they went, but the team’s priority was to craft the story so that a nonexpert—like members of the judging team who were not battery guys—could understand it. As the proposal stood, it failed to meet this standard.

  More problematically, the proposal seemed actually to ignore some provisions of the FOA. The FOA had stipulated a serious commitment to applied science. Madia judged the appropriate balance at about 60 percent research and the rest development and deployment. The Argonne team had proposed an 80 percent emphasis on basic research—clearly too much.

  He raised a couple of other points—there were too many “whats” and not enough “hows”; each time the proposal said the team intended to do something, it should provide an example of how it would be done. Madia was troubled. The previous summer, he had seen a preliminary draft and said much the same. “You’re no further than you were nine months ago,” he now told the group.

  Chamberlain’s face went crimson. He was “mad as hell,” Coghill said later. After a while, he collected himself. But Madia did not back down. There had to be “significant revisions,” he said.

  In its current shape, the proposal did not merit scoring. And he would not do it. The best course was to start rewriting.

  Chamberlain e-mailed Isaacs: I think Madia is trying to manipulate us and we need to understand his motives, he wrote. Isaacs knew that Chamberlain’s team was “sore as hell.” He responded by spelling out explicitly what until then had gone unstated: Madia was an adviser. Chamberlain was in charge.

  In the hours after the thrashing, Isaacs dined with Madia, who openly questioned Chamberlain’s readiness for the project. The leader of such a proposal team was analogous to an air traffic controller, he said. He had to keep thirty-one planes in the air, ensure that they did not crash into one another, and hold them in their slots when one wished to land. Such navigation required alacrity. Unlike airplane pilots, who always followed instructions, Chamberlain had to preside over autonomous players. The battery guys alone included Stanford’s Yi Cui, Berkeley’s Venkat Srinivasan, and MIT’s Gerd Ceder and Yet-Ming Chiang. They were all stars in their own right. The Argonne leadership—Mark Peters and Isaacs himself—had to be handled as well, as did the industry players. Could Chamberlain do all that? Madia had his doubts.

  Isaacs thought that Chamberlain possesse
d the raw skills, and told Madia so. The situation was not as dire as he claimed. But there was something to Madia’s message: if the proposal was going to come together, Chamberlain did need to up his game. He was exceptionally smart; when he put his mind to something, he could dig extremely deep into the science. He was technologically and entrepreneurially savvy—in the IP department, no one was better at putting together a deal. But while those were admirable traits, they did not add up to exceptional leadership—not when it came to something of the scale of the Hub proposal. In the Battery Department, Chamberlain had done well at managing a couple of stars—Thackeray and Amine. But as Madia said, there were many more heavyweights on the Hub team. From a tactical perspective, Chamberlain had to figure out how to nudge these personalities to work together—to work for him. If there was an inflection point on the way to this new and improved Chamberlain, it would have to be now.

  Madia asked Isaacs whether he could abandon protocol, under which he was to stand aloof from the proposal writers, and converse face-to-face with Chamberlain. Chamberlain had asked to see him, he said—perhaps it would help.

  “Of course,” Isaacs said.