by Jon Gertner
Five
SOLID STATE
In technology and business—and in men’s lives, too—the war cleaved the past from the future. Shockley, for instance, would not be returning to the Labs’ West Street offices in Manhattan. Kelly’s postwar plans for him, as well as for Bell Labs, depended on a move across the Hudson River. For the past five years, in between all of his wartime work, Kelly had devoted large blocks of time to orchestrating the construction of an immense complex of buildings in the wooded hills of New Jersey. Situated in a neighborhood called Murray Hill that was set within the quiet suburban towns of New Providence and Berkeley Heights, the new office was to be located about twenty-five miles from New York. “It was decided in the mid-thirties to initiate a gradual exodus from the city,” Kelly recalled some years later, explaining that the aging labs at West Street had been deemed “non-functional” by that point. A 225-acre tract in Murray Hill had already been acquired. In July 1930 Frank Jewett, the Labs president, had called a public meeting near the future location to announce the land acquisition and describe a plan to build a research facility.1 Those plans had been put on hold until the worst years of the Depression had passed, and in 1938 the idea of a grand suburban laboratory was revived by Jewett, then–vice president Oliver Buckley, and research chief Kelly.
Some of their reasons were purely scientific. For years Bell Labs had been operating small satellite facilities at far-flung locations around New Jersey—near the shore in the towns of Holmdel and Deal, for instance, and in the forested hills near the North Jersey town of Whippany. Long-wave and shortwave radio researchers at those outposts needed distance from the interference of New York City (and from one another) to do proper research and measurements. Murray Hill was put in a similar context: A move to the suburbs would allow the physics, chemistry, and acoustics staff to conduct research in a location unaffected by the dirt, noise, vibrations, and general disturbances of New York City.
It was also true, however, that Bell Labs was growing at such a clip it had simply overrun its facilities. In a private memo to Jewett, Buckley noted that “these buildings at Murray Hill [would] secure relief from an unsatisfactory and hazardous housing situation in New York City.” A new laboratory, in other words, would allow for the ultimate consolidation of the organization’s research staff, which was now spread out in a number of rented buildings and even in the basement of 463 West Street. In his memo, Buckley could have added another reason for the move, but it would have been unnecessary: The new site was close to his home and Jewett’s. Buckley lived a few minutes away in the town of Maplewood; Jewett was in Short Hills, as were Kelly and Davisson. Shockley lived in nearby Madison as well. The executives were building a new laboratory in their own backyard.
With a price tag of about $4.1 million, the new building was not conceived as an ordinary laboratory.2 By the late 1930s, Murray Hill had become a research and development project in itself, designed with meticulous care by Buckley and Kelly and a number of other engineers so that it would not only relieve Bell Labs’ congestion problems but would organize its scientists in an unusual configuration that could be expanded on a far grander scale in future years. Kelly, Buckley, and Jewett were of the mind that Bell Labs would soon become—or was already—the largest and most advanced research organization in the world. As they toured industrial labs in the United States and Europe in the mid-1930s, seeking ideas for their own project, their opinions were reinforced. They wanted the new building to reflect the Labs’ lofty status and academic standing—“surroundings more suggestive of a university than a factory,” in Buckley’s words, but with a slight but significant difference. “No attempt has been made to achieve the character of a university campus with its separate buildings,” Buckley told Jewett. “On the contrary, all buildings have been connected so as to avoid fixed geographical delineation between departments and to encourage free interchange and close contact among them.”3 The physicists and chemists and mathematicians were not meant to avoid one another, in other words, and the research people were not meant to evade the development people.
By intention, everyone would be in one another’s way. Members of the technical staff would often have both laboratories and small offices—but these might be in different corridors, therefore making it necessary to walk between the two, and all but assuring a chance encounter or two with a colleague during the commute. By the same token, the long corridor for the wing that would house many of the physics researchers was intentionally made to be seven hundred feet in length. It was so long that to look down it from one end was to see the other end disappear at a vanishing point. Traveling its length without encountering a number of acquaintances, problems, diversions, and ideas would be almost impossible. Then again, that was the point. Walking down that impossibly long tiled corridor, a scientist on his way to lunch in the Murray Hill cafeteria was like a magnet rolling past iron filings.
MURRAY HILL’S FIRST BUILDING—Building 1, as it was eventually known—officially opened in 1942.4 Inside it was a model of sleek and flexible utility. Every office and every lab was divided into six-foot increments so that spaces could be expanded or shrunk depending on needs, thanks to a system of soundproofed steel partition walls that could be moved on short notice. Thus a research team with an eighteen-foot lab might, if space allowed, quickly expand their work into a twenty-four-foot lab. Each six-foot space, in addition, was outfitted with pipes providing all the basic needs of an experimentalist: compressed air, distilled water, steam, gas, vacuum, hydrogen, oxygen, and nitrogen. And there was both DC and AC power. From the outside, the Murray Hill complex appeared vaguely H-shaped. Most of the actual laboratories were located in two long wings, each four stories high, which were built in parallel and were connected by another wing. Set back a thousand feet from the street, Building 1—finished in limestone and buff-colored brick and roofed with copper sheeting that in a few years’ time would acquire a green patina—was fronted by a vast apron of lawn. Buckley and Kelly wanted quiet, and quiet is what they got. The roads on all four sides of the complex were lightly traveled, and behind the building were hundreds of acres of protected forest set in a county reservation of rolling hills. In the mornings, before the nine hundred or so scientists and technical assistants arrived, the massive building set amid the greenery had a grand and rarefied hush. The Bell Labs executives had not only built a new lab; they had built a citadel.
As American industry made the transition to postwar work, scientists from around the country made a pilgrimage. “Its use in our military research activities during the war years demonstrated its great worth,” Kelly said with pride in the mid-1940s, as if it were another Bell Labs invention, which in fact he thought it was. “It has attracted so much attention that representatives of more than eighty industrial laboratories have visited it and have obtained detailed information from us about its special functional features,” he added. Many laboratories patterned after Murray Hill were now under construction in the United States; European research managers and architects had come to study the building, too. Meanwhile the location, and not just the building, had drawn the interest of other companies, leading a number of industrial labs to locate nearby. One newspaper dubbed the New Jersey suburbs “research row.”
Bell Labs was Kelly’s shop now—officially, that is. In 1944 Frank Jewett had elevated himself to Bell Labs chairman, a mostly honorary position, as the war and Washington politics consumed his attention. Buckley had meanwhile assumed the presidency, a rank that put him on a busy circuit of speaking engagements around the country while also forcing him to tend to AT&T corporate obligations in New York City. Kelly became executive vice president, giving him complete control over day-to-day operations. One of his first acts was reorganizing the Murray Hill staff one day in July 1945. On that morning, several supervisors were demoted, while men younger and better versed in solid-state physics, such as Bill Shockley, were promoted. Dean Woolridge told the science historian Lillian Hoddeson that “there w
ere rumors that Kelly was reorganizing for peacetime and everybody knew something was going to happen when he announced a big meeting.” Many, if not all, of the supervisors in the research department were called together in a large room with Kelly. “He sat up there at the head of the room and he read off ‘From now on thou shalt do this and thou shalt have this particular group and you’re going to move over here and do this kind of work and you’re going to do this.’” Kelly had worked out the new organizational structure in minute detail. The meeting took most of the day.
Essentially Kelly was creating interdisciplinary groups—combining chemists, physicists, metallurgists, and engineers; combining theoreticians with experimentalists—to work on new electronic technologies. But putting young men like Shockley in a management position devastated some of the older Labs scientists. Addison White, a younger member of the technical staff who before the war had taken part in Shockley’s weekly study group, told Hoddeson he nevertheless considered it “a stroke of enormously good management on Kelly’s part.” He even thought it an act of managerial bravery to strip the titles from men Kelly had worked with for decades. “One of these men wept in my office after this happened,” White said. “I’m sure it was an essential part of what by this time had become a revolution.”5
No documentation, and no private papers, have ever surfaced to fully explain Kelly’s rationale.6 Was he at this point looking for an actual invention that he could now clearly articulate, or was he just acting on a hunch that something useful was waiting to be found, and that if it weren’t something big, then surely combining men of such talent in a state-of-the-art building would produce a breakthrough? There exists only one true founding document of the revolution: an authorization form that Kelly signed to fund the new groups he had organized. Under Bell System protocol, work at Bell Labs had to be billed to either AT&T, Western Electric, or the local operating companies like Pacific Telephone. Different types of research were classified as a “case,” and each case was in turn approved (or, rarely, challenged) by corporate management. Vacuum tube development, for instance, had its own case number, as did basic physics research. This was one reason why Kelly had quashed Fisk’s swimming pool requisition: It could hardly be justified under an existing case number, and even if it could, a new case for building a swimming pool was bound to be met at AT&T headquarters with a skepticism rivaling Kelly’s own.
On June 21, 1945, Kelly had signed off on Case 38139. “A unified approach to all of our solid state problems offers great promise,” he wrote. “Hence all of the research activity in the area of solids is now being consolidated in order to achieve the unified approach to the theoretical and experimental work of the solid state area.” The point of the new effort, in case it was hard to see through the jargon, was “the obtaining of new knowledge that can be used in the development of completely new and improved components” of communications systems. At the end of the six-page document, Kelly noted that he did not anticipate that the solid-state work would bring immediate results to the telephone business. On the other hand, he reasoned, the research was “so basic and may well be of such far-reaching importance” to the business that it was imperative that the phone company supply the funding. The initial cost of the program was put at $417,000, most of which comprised salaries for the group members. The authorization was signed by Harvey Fletcher, Jim Fisk, and Kelly himself. It was billed to AT&T.7
One by one, meanwhile, the men who had left West Street for some military reason or another—for the battlefields, for work in Washington, for a stint in Whippany or at the biscuit building in Manhattan to design radar sets—were filtering back. Walter Brattain had spent the first half of the war working in Washington on submarine detection and the second half at Bell Labs working on other matters of military engineering. In Kelly’s new order, he would be joining a solid-state research group led mainly by Bill Shockley. The notebook Brattain had used to chart his semiconductor experiments before the war—notebook number 18194—had its last entry in West Street on November 7, 1941. Four years later, in the new Murray Hill building, Brattain picked it up again and opened to page 40. “The war is over,” he wrote.8
IN LATER YEARS, it would become a kind of received wisdom that many of the revolutionary technologies that arose at Bell Labs in the 1940s and 1950s owed their existence to dashing physicists such as Bill Shockley, and to the iconoclastic ideas of quantum mechanics. These men could effectively see into the deepest recesses of the atom, and could theorize inventions no one had previously deemed possible. More fundamentally, however, the coming age of technologies owed its existence to a quiet revolution in materials. Indeed, without new materials—that is, materials that were created through new chemistry techniques, or rare and common metals that could now be brought to a novel state of ultrapurity by resourceful metallurgists—the actual physical inventions of the period might have been impossible. Shockley would have spent his career trapped in a prison of elegant theory.
A few of the scientists at Bell Labs grasped the fact earlier than others. Just as the men in Thomas Edison’s old laboratory would tinker with animal hoofs and horsehair in their inventions, for the first three or four decades of the phone system its engineers worked mostly with everyday substances—wooden blocks hewn from the finest bird’s-eye maple to mount switching equipment, or a natural latex known as gutta-percha for waterproofing cables. Increasingly, however, new ideas, and new projects, ran up against the limitation of nature’s design. “We had such specific requirements that ordinary raw materials had an agonizing time meeting them,” William Baker, who joined the Labs as a chemist just before World War II, explained.9 The solution, as Baker described it, was to literally create new types of matter. When the Labs’ chemists needed to help design impermeable underwater cables, for example, one possibility was gutta-percha. But gutta-percha had drawbacks, including its extreme expense. For an undersea cable to Catalina Island, off the coast of California, the Labs’ chemists began looking for an alternative. Natural rubber was considered too soft in its pure state, and when it was treated with sulfur to toughen it—that is, to “vulcanize” it—the men had to address the problem that sulfur corrodes copper and would undoubtedly degrade the vital wires within the undersea cable. Only after the chemists determined that they could purify the rubber in a complex manner and then create fine silica flour as an insulator could the cable go into production.
Those working with metals wrestled with challenges similar to those working with rubber and plastics. During the Labs’ early days, while Mervin Kelly was running the tube shop in lower Manhattan, metallurgists began to focus on whether special coatings on metal filaments inside the vacuum tubes could vastly improve their performance or durability. The work required them to delve into the uses of obscure elements and alloys, and to conceive—successfully—of arcane processes to heat and cool their mixtures. A similar example arose when Labs scientists tried to improve the thin diaphragm in a phone receiver—the metal disc that vibrates in response to a speaker’s voice. They ultimately created something called Permendur, an alloy of cobalt and iron that was spiced with around 2 percent of the element vanadium. But the metallurgists soon realized that Permendur was only part of the solution to finding a better diaphragm. If the ingredients in the alloy weren’t pure—if they happened to contain minute traces of carbon, oxygen, or nitrogen, for instance—Permendur would be imperfect.10 “There was a time not so long ago when a thousandth of a percent or a hundredth of a percent of a foreign body in a chemical mixture was looked upon merely as an incidental inclusion which could have no appreciable effect on the characteristics of the substance,” Frank Jewett, the first president of the Labs, explained. “We have learned in recent years that this is an absolutely erroneous idea.”11
It was understandable, then, for an engineer or scientist to regard the era of postwar electronics much like many people at Bell Labs did: New devices were waiting to be discovered through the combination of basic research and the i
ntellectual force of physicists and engineers working together. The world of electronics could move forward through new ideas and the meticulous work of development engineers. “I think that there are vistas ahead that [are] as large or larger than the past,” Mervin Kelly declared in May 1947.12 A more cautious view maintained that the potential new age depended on finding a solution to a cosmic puzzle. Progress, in both technology and business, depended on new materials, and new materials were scattered about the earth in confusion. There were substances that might be useful on their own or combined into compounds—that alone comprised an infinite number. And each of these potential substances could, in turn, be rendered in a laboratory into an almost unimaginable range of purities. One might conceivably find the right material, with the right level of purity, for a new device. Or one might not.
THE PURSUIT OF PURITY at Bell Labs went further than Frank Jewett or Mervin Kelly ever imagined. In 1939, the year Shockley and Brattain had tried and failed to make a solid-state amplifier out of copper oxide, the chemists and metallurgists at the Labs had already begun looking closely at the class of materials called semiconductors. In particular they focused their inquiries on silicon. Already physicists had concluded that many semiconductors exhibited a unique set of behaviors related to their atomic structure. Atoms have a nucleus packed full of protons and neutrons that are surrounded by bands of vibrating electrons. In a good conductor—copper, for instance—the band farthest from the nucleus has only one or two electrons, which means this band is mostly empty. The outermost electrons are often free to bounce around and move to neighboring copper atoms. In a good insulator—glass, for instance—the opposite holds true; the band farthest from the nucleus has seven or eight electrons, which means it is mostly full. The electrons are therefore held in fixed positions. These contrasts translate into differences in how the materials conduct electric current, which might be thought of as a “flow” of electrons through a solid material. In a conductor, electrons move freely. In an insulator, they don’t. As the Bell Labs researchers would describe it, the electrons in an insulator essentially “act as a rigid cement” to bind together the atoms in the solid.