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

by Steve LeVine

  Goodenough’s boyhood did not suggest the warm, amusing, and self-assured adult to come. Suffering from dyslexia at a time when it was poorly understood and went untreated, Goodenough could not read at Groton, understand his lessons, or keep up in the chapel. Instead, he occupied himself in explorations of the woods, its animals and plants. Somehow everything came together. He went on to thrive at Yale, from which he graduated summa cum laude in mathematics, then by happenstance fell into science: after World War II, Goodenough, by then a twenty-four-year-old Army captain posted in the Azores Archipelago off the coast of Portugal, received a telex ordering him to Washington, D.C.—educators had stumbled on unspent budget money and advocated using it to send twenty-one returning Army officers through graduate studies in physics and math. Goodenough had taken almost no science as an undergrad but, for reasons obscured by time, a Yale math professor had added his name to the group. So he found himself at the University of Chicago, studying physics under professors Edward Teller, Enrico Fermi, and others. As Goodenough registered for preliminary undergraduate classes, necessary to catch up with the others, a professor remarked, “I don’t understand you veterans. Don’t you know that anyone who has ever done anything significant in physics has already done it by the time he was your age?”

  But it turned out that Goodenough had an intuition for physics. After obtaining his doctorate in 1952, he went to work at MIT’s Lincoln Laboratory, which the U.S. Air Force had funded the year before to create the country’s first air defense system. His team was told to invent a system of computer memory, a vital component of the envisioned air defense, which was to be called SAGE. At the time, computers comprised enough vacuum tubes to fill “the space of a large dance hall,” in Goodenough’s words, and had infernally slow memories.1 Some thought the task impossible because of the physical limits of the ceramic material with which the team was working. Three years later, the lab unveiled an invention that they called “64 x 64 bit magnetic memory,” a triumph that, in addition to helping to enable SAGE, became the foundation of later computer memory systems. For Goodenough, more advances followed, including the “Goodenough-Kanamori rules,” which became a standard for how metal oxide materials behave at the atomic scale, another building block of future computers.

  Politics intruded—a U.S. senator named Mike Mansfield pushed through a law requiring that any research financed by the Air Force have an Air Force application. By now, Goodenough was fixated on finding a scientific answer to the OPEC-led energy crisis, which seemed to be the largest problem facing the country. But he was told to try something else—given the Mansfield law, the subject was the responsibility not of the Air Force but of the national labs.

  For Goodenough, it was time to move on. A friend sent word of an opportunity across the Atlantic. Oxford University required a professor to teach and manage its inorganic chemistry lab. Goodenough was surprised to be selected given that he was not a chemist and in fact had completed just two college-level chemistry courses. He was lucky a second time to be chosen for a job for which he was underqualified, on paper.

  • • •

  Goodenough was a tough professor. An early student of his at Oxford recalled a physics course that started with 165 students. After a stern Goodenough lecture, she was one of just 8 to return for the second class.2 Goodenough was equally exacting in the lab. After MIT, he was on the hunt for big advances in solid state chemistry, a field known for creating the kinds of materials that go commercial. Among the first on his list of targets was Stan Whittingham’s recently published breakthrough on the lithium battery.

  For six decades, zinc carbon had been the standard battery chemistry for consumer electronics, having eclipsed lead oxide, which was too bulky and heavy for small devices. Whittingham’s brainchild was a leap ahead of zinc carbon—powerful and lightweight, it could power portable consumer electronics such as tape recorders. If it worked. But basic physics got in the way. The same electrochemical reactions that enabled lithium batteries also made them want to explode: the voltage would run away with itself, a cell would ignite, and before you knew it the battery was spitting out flames. But you seemed no better off if you played it safe and used other elements—you’d find that they slowly fell apart on repeated charge and discharge.

  Goodenough thought he could create a more powerful battery than Whittingham’s. Much of invention, he said, involves shifting your mind-set, something many scientists either refused or simply could not do. The Exxon man’s battery relied on a sulfide electrode; Goodenough turned to another family of compounds—metal oxides, a combination of oxygen and a variety of metal elements. In his judgment, oxides could be charged and discharged at a higher voltage than Whittingham’s creation, and thus produce more energy. But there was also the matter of getting sufficient lithium to intercalate, the action that created electricity—pulling lithium from a cathode, in this case made of metal oxide, and sending it into a shuttling motion between the electrodes. The more lithium that could be shuttled, the more energy the battery would produce. But it seemed axiomatic that you could not remove all the lithium, because that would leave the cathode virtually hollowed out, and it would fall in on itself. So could any of the oxides manage to hold up under this abuse? And if so, which one, and what was the magic proportion of lithium that could be pulled out?

  Goodenough directed two postdoctoral assistants to methodically work their way through structures containing a group of oxides; he asked them to find out at what voltage lithium could be extracted from the oxides, which he expected to be much higher than the 2.2 volts Whittingham was using, and to determine how much lithium could be intercalated in and out of the atomic structure before it collapsed. Their answer was half—about 50 percent of the lithium could be pulled from the cathode at 4 volts before it crumpled, which was plenty for a powerful, rechargeable battery. Of the oxides they tested, the postdocs found that cobalt was the best and most stable for this purpose.

  In 1980, four years after Goodenough arrived at Oxford, lithium-cobalt-oxide was a breakthrough even bigger than Ford’s sodium-sulfur configuration. It was the first lithium-ion cathode with the capacity to power both compact and relatively large devices, a quality that made it far superior to anything on the market. Goodenough’s invention changed what was possible: it enabled the age of modern mobile phones and laptop computers. It also opened a path to the investigation of a potential resurrection of electric vehicles.

  Over the years, Goodenough would attract a constellation of bright people to his lab, researchers who often had their best professional years with him. It was not that Goodenough himself did any of the hands-on experimentation—the postdoctoral assistants and researchers he attracted were actually at the bench. Goodenough could be stern, but the atmosphere of big expectation he created drove them to do exceptional work. And he talked them through their projects. One of these researchers was a young South African who arrived in 1981 with a curious idea about gemstones.

  6

  The Double Marathoner

  The Comrades Marathon extends to the South African port of Durban from Pietermaritzburg, twenty-eight miles inland and three thousand feet lower in altitude. The first time that Mike Thackeray ran the race, in 1968, he finished in ten hours and three minutes, just under the eleven-hour cutoff for the slowest participants. Determined to do better, he ran it again. And again. In 1976, entering the race for the fourteenth time, Thackeray took fourth place with a time of 6:32. His discipline had paid off.

  Thackeray was the lead inventor of Argonne’s NMC technology, a descendant of the lithium-cobalt-oxide cathode pioneered by Goodenough and the formulation that had beguiled Wan Gang. Thackeray’s office was situated within the main Battery Department suite, two doors down from his boss Chamberlain. Long halls lined with linoleum and pale green brick walls gave Building 205 a lingering feel of the 1950s. A handwritten sign taped to a coffee brewer requested that drinkers leave behind thirty cents a cup.

/>   Two portraits decorated the walls in Thackeray’s office—an 1861 etching of the nineteenth-century physicist Michael Faraday and a sketch of the astronomer William Herschel, who in 1781 discovered Uranus. Thackeray received them as gifts in his youth in South Africa. His mind returned often to his native land, which seemed to speak the most for his soul. Few knew it, he would say, but for a short time almost four decades before, South Africa was one of the great centers of battery thinking.

  In Pretoria in the late 1970s, Thackeray, in shaggy, blondish hair and long sideburns, did his Ph.D. under a crystallographer named Johan Coetzer. One day, Coetzer walked into the lab and announced a new project. They were going to “do some stuff in the energy field.” The Yom Kippur War between Israel and its Arab neighbors had triggered an energy crisis and the Western world was seeking a way around Middle East oil. Coetzer thought one answer was the advancement of batteries and he told Thackeray that that was where they would focus their work. The effort was challenged from the beginning because of South Africa’s system of apartheid, to which the world had responded with economic sanctions. No one outside the country would collaborate with them. To avert international trouble, they had to cloak their work in secrecy and communicate using code words. The smokescreen did not seem to matter much since neither Coetzer nor Thackeray knew anything about energy storage. But their fresh eyes turned out to be advantageous. Approaching the field laterally, “uncontaminated by how other scientists were looking at the world,” as Thackeray put it, they found insights into high-temperature batteries, the breakthrough reported by Ford and Stanford. The early result was the Zebra, South Africa’s own molten battery. Corporate funding quickly followed, a respectable achievement when you recalled their modest start.

  Considering the Zebra, Thackeray thought there still must be a way to do better and at the same time move ahead of John Goodenough’s blockbuster advance in 1980. The Zebra and other molten batteries, operating at 300 degrees Celsius, were unsafe, inside a car anyway. As for Goodenough’s room-temperature formulation of lithium-cobalt-oxide, it was an improvement but still expensive if you thought of using it in electronic devices.

  In physics, there is a structure called spinel. These structures have considerable advantages. They are abundant and therefore cheap. They have an appealing three-dimensional structure resembling a crystal. And spinels are inherently stout—sturdier, for instance, than the layered structure of Goodenough’s lithium-cobalt-oxide electrode. Goodenough had been instructing his lab assistants to put half the lithium in motion between the cathode and the anode; but Thackeray wondered whether all the lithium could be pulled in and out of a spinel cathode. If he could do so without the cathode’s collapsing, spinel would be less expensive and potentially much more powerful than the lithium-cobalt-oxide.

  The particular spinel that interested Thackeray was iron oxide. Ordinarily, we know iron oxide as rust—it is what happens when you leave your bicycle out in the rain. But for battery scientists, iron oxide is also a spinel, lending it special characteristics. In South Africa, Thackeray had already successfully shuttled lithium in and out of iron oxide working at the same high temperatures as the Ford researchers. He had a hunch that iron oxide might also cooperate at room temperature, which would make it much more practical.

  South Africa was disconnected geographically as well as politically. It was almost as far as you could be from the intellectual hubs of the United States and Europe. That being the case, it was almost expected that any self-respecting young South African scientist would spend a year or so abroad. Thackeray decided that he wanted to use his own sabbatical to test his ideas with spinel. And to do so with the leading figure of the day—Goodenough.

  Thackeray wrote to Oxford. Goodenough responded immediately: he lacked money to support the younger man, but if that didn’t bother Thackeray, he would be pleased to play host. Thackeray, owed his lab-funded obligatory time abroad, required no outside funding. So, at thirty-one, he along with his wife and daughter packed for a fifteen-month postdoctoral assistantship in England.

  • • •

  Thackeray wandered the Oxford campus. As he looked around, he recalled his father’s stories of undergraduate study. Generations of Thackerays had attended Cambridge—his father, Andrew David Thackeray, who rose to be a leading astronomer; his grandfather, Henry St. John Thackeray, a biblical scholar who went on to teach at the university; and of course the novelist William Makepeace Thackeray, a fifth cousin once removed. Both Thackeray and his brother had elected to remain in South Africa for university, and he sensed himself unequal to England’s great academic institutions. The intimidation was not just Oxford, but Goodenough himself. Thackeray found the older man’s intelligence almost overpowering. He had never heard anything quite like the professor’s resonant hoot, an unusual chortle that Goodenough often let fly. By comparison, Thackeray regarded himself as an ordinary “bush chemist from Africa.”

  He felt out of place for yet another reason—because of his country’s medieval political system, he was certain that those he encountered, while saying nothing, must be harboring repulsive thoughts about him and his family. But Goodenough himself plainly did not hold Thackeray or his family personally responsible for the sins of their country. Thackeray would have his own, very different proving ground: the lab.

  • • •

  Thackeray had brought samples of magnetized iron oxide spinel with him from Pretoria. He planned to intercalate lithium in and out of them at room temperature and thus demonstrate that iron oxide spinel could be a powerful new battery material, one commercially superior to the cobalt formulation.

  Goodenough dismissed Thackeray’s hypothesis. It violated physics—spinels resemble semiprecious gemstones and, structurally speaking, you could not move lithium in and out of a gemstone, the older man reminded Thackeray. Its physical structure, unlike cobalt oxide, would block any such attack.

  Thackeray recalled how, despite Goodenough’s skepticism, he had already managed the deed at high temperatures back in South Africa. This was an issue of merely lowering the temperature.

  “Well, you are welcome to try,” Goodenough said. “But you’ll want to look around the lab for other stuff to do.” Then he left for holiday in India.

  Two weeks later, Goodenough returned. “I intercalated the lithium,” Thackeray said.

  “What?”

  The older man took Thackeray into his office and listened as he explained how, using a magnetic stirrer, an automated device for mixing chemicals, he had combined lithium with iron oxide at room temperature. Thackeray observed an immediate encouraging sign—the iron oxide fell away from the stirrer, showing that it had lost its magnetic qualities. That suggested that the spinel had ingested the lithium. Yet this was still not conclusive evidence of intercalation. There had to be more if Thackeray was to state definitively that this was the case. He smeared the concoction onto a glass slide. Now he shot X-rays through it. When you take such pictures you get a spectrum of peaks rather than an image. The trick is to infer the precise structure of the compound from the pattern of the peaks. This skill, the knowledge of X-ray diffraction, is known as crystallography.

  Thackeray had shot two X-rays—one prior to the experiment and one after. Comparing them, he noticed “striking differences” in both the position of the peaks and their relative intensity. Something had happened. If Goodenough was right, and the lithium found no entry into the iron oxide structure, the X-ray patterns would be identical. But the peaks had noticeably changed—the lithium had intercalated into the iron oxide. Iron oxide spinel could be fashioned into a lithium-ion electrode.

  In fact, as Goodenough had postulated, there wasn’t space for the lithium in the spinel. What Thackeray had shown was that the spinel had an unexpected quality of hospitality—when you moved lithium in, the iron ions shifted around to accommodate it. They made extra space. The spinel experienced a “phase change,” absorbing the iron and t
ransforming into a slightly different material resembling rock salt. Like Goodenough’s lithium-cobalt-oxide breakthrough a year earlier, Thackeray had conceived of a way to significantly improve on the energy density of zinc carbon batteries. Goodenough was surprised and enthusiastic—Thackeray’s idea proved a new principle and, from the standpoint of cost, was potentially better than his own brainchild.

  Yet he had not created a practical battery material—a workable cathode—which was the objective. There was a problem, Goodenough said. Examining the data, he detected a blockage in the spinel. The iron oxide wasn’t providing a clear path for sufficient lithium to enter and find a home in the structure before being shuttled out in the charge-recharge cycle. A less-cluttered channel was needed if the material was to be truly useful and not a mere novelty.

  Perhaps the problem was the type of spinel they were using. A different sort—possibly manganese oxide, which he called by its scientific notation, LiMn2O4—could lift the logjam and allow the lithium into the right places. Goodenough had intimate knowledge of manganese oxide from his MIT days because his team had used it in their computer memory experiments. He suggested that they swap oxides. Manganese spinel, LiMn2O4, could prove the path to a cheaper cathode.