Genius: The Life and Science of Richard Feynman
Making a bomb was not like making a theory of quantum electrodynamics, where the ground had already been mined by the greatest scientists. Here the problems were fresh, close to the surface, and therefore—this surprised Feynman at first—easy. Beginning with the issues raised by the first indoctrination lectures, he produced a string of small triumphs, gratifying by contrast with the long periods of wandering in the dark of pure theory. There were compensating difficulties, however.
“Most of what was to be done was to be done for the first time,” an anonymous ghostwriter of the bomb’s official history wrote afterward. (The ghostwriter was Feynman, called to this unaccustomed service by his former department head, Harry Smyth.) Struggling to sum up the problems of theoretical science at Los Alamos, he added “untried,” and then “with materials which were for a long time practically unavailable.” Materials—he could not bring himself to write uranium or plutonium after the euphemistic years of tubealloy and 49. The wait for tubealloy had been agonizing, for the theorists no less than the experimenters. More mundane materials could be requisitioned—at the laboratory’s request Fort Knox delivered two hemispheres of pure gold, each the size of half a basketball. Feynman, giving Smyth a tour one day, pointed out that he was absently kicking one of them, now in use as a doorstop. A request for osmium, a dense nonradioactive metal, had to be denied when it became clear that the metallurgists had asked for more than the world’s total supply. In the cases of uranium 235 and plutonium, the laboratory had to wait for the world’s supply to be multiplied a millionfold.
For now the only knowledge of these materials came from experiments on quantities so tiny as to be invisible. The experiments were expensive and painstaking. Even getting an early measurement of plutonium’s density challenged the team at Chicago. The first dot of plutonium did not arrive at Los Alamos until October 1943. Trials with more comfortable quantities would have to wait; in the event, just one full-size experiment would be possible. Most questions would have to be answered with pencil and paper. It soon became clear that theory at Los Alamos would be performed on a high wire without a net. The theoretical division was small, just thirty-five physicists and a computing staff, charged with providing analysis and prediction for all the much larger practical divisions: experimental, ordnance, weapons, and chemical and metallurgical. Analysis and prediction—what would happen if… ? Theorists at Los Alamos had dispensed with the luxury of contemplating simple mysteries—the way a single atom of hydrogen emits a single packet of light in such and such a color, or the way an idealized wave might travel through an idealized gas. The materials at hand were not idealized, and the theorists, no less than the experimenters, had to poke about in the rubble-strewn territory of nonlinear mathematics. Crucial decisions had to be made before the experimenters could conduct trials. Feynman, in his anonymous account, listed the main questions:
How big must the bombs be (the imploding sphere of plutonium or the gun device in the case of uranium)? What would be the critical mass and the critical radius for each material, the dimensions beyond which a chain reaction would sustain itself?
What materials would best serve as tamper, a surrounding liner that would reflect neutrons back into the bomb? The metallurgists had to begin the work of fabricating tamper long before a true test was possible.
How pure would the uranium have to be? On this calculation rested a decision to build or not build an enormous third stage in the isotope-separation complex at Oak Ridge.
How much heat, how much light, how much shock would a nuclear explosion create in the atmosphere?
The Battleship and the Mosquito Boat
They occupied a two-story green-painted box called T building (T for theoretical), which Oppenheimer made his headquarters and the laboratory’s spiritual center. He placed Hans Bethe, Cornell’s famous nuclear physicist, in charge. The corridors were narrow, the walls thin. As the scientists worked, they would hear from time to time Bethe’s booming laughter. When they heard that laugh they suspected that Feynman was nearby.
Bethe and Feynman—strange pair, some of their colleagues thought, a pedantic-seeming German professor and a budding quicksilver genius. Someone coined the nicknames “Battleship” and “Mosquito Boat.” Their collaborative method was for Bethe to plow solidly ahead, a determined giant, while Feynman buzzed back and forth across his bow, gesticulating, yelling in his scabrous New York accent, “You’re crazy” and “That’s nuts.” Bethe would respond calmly in his slow professorial way, working his way through the problem analytically and explaining that he was not crazy, Feynman was crazy. Feynman would consider and pace back and forth, and finally through the partitions the other scientists would hear him shout back, “No, no, you’re wrong.” He was reckless where Bethe was careful, and he was just what Bethe was looking for, someone who would perform the severest and most imaginative criticism, who would find flaws before an idea went too far. Challenges and fresh insights came easily from Feynman. He did not wait, as Bethe did, to double-check every intuitive leap. His first idea did not always work. His cannier colleagues developed a rule of thumb: If Feynman says it three times, it’s right.
Bethe was a natural choice as leader of the theoretical division. His sweeping three-article review of the state of nuclear physics in the thirties had established him as the authoritative theorist in that field. As Oppenheimer well knew, Bethe had not just organized the existing knowledge of the subject but had calculated or recalculated every line of theory himself. He had worked on probability theory, on the theory of shock waves, on the penetration of armor by artillery shells (this last paper, born of his eagerness in 1940 to make some contribution to the looming war, was immediately classified by the army so that Bethe himself, not yet an American citizen, could not see it again). His explanation in 1938 of the thermonuclear fires that light the sun would win him the Nobel Prize. Since arriving at Cornell in 1935 he had made it one of the new world centers in physics, as Oppenheimer and Ernest O. Lawrence had done for Berkeley.
Oppenheimer wanted him badly and strained to persuade him that the atomic bomb was practical enough to draw him from the MIT Radiation Laboratory, where he had begun to make a contribution in 1942. (When Bethe agreed, the news was sent to Oppenheimer by a prearranged code: a Western Union kiddiegram.) Bethe’s friend Edward Teller had pressed hard for his participation. No one but Teller was now surprised when Oppenheimer appointed Bethe, the sturdy pragmatist, to head the theoretical division, to nurse the egos and the prodigies, to run the most eccentric, temperamental, insecure, volatile assortment of thinkers and calculators ever squeezed together in one place.
Bethe had learned his physics all across Europe: first at Munich, where he studied with Arnold Sommerfeld, a prodigious producer of future Nobel Prize winners, and then at Cambridge and Rome. At Cambridge, Dirac’s lectures on the new quantum mechanics held center stage, but Bethe quit attending after discovering that Dirac, having perfected his formulation of the subject, was simply reading his book aloud. At Rome, where he was the first foreign student of physics in the university’s history, the attraction was Fermi. For a short time they worked together closely, and Bethe acquired from him a style that he called “lightness of approach.” His first great teacher, Sommerfeld, had always begun work on a problem by writing down a formalism selected from a heavy arsenal of mathematical equipment. He would work out the equations and only then translate the results into an understanding of the physics. By contrast, Fermi would begin by gently turning a problem over in his mind, by thinking about the forces at work, and only later sketching out the necessary equations. “Lightness” was a difficult attitude to sustain in a time of abstract, unvisualizable quantum mechanics. Bethe combined the physicality of Fermi’s attitude with an almost compulsive interest in computing the actual numbers that an equation entailed. That was far from typical. Most physicists could happily string equations down a page, working out the algebra without keeping in mind a sense of real quantities, or ranges of quantiti
es, that a symbol might represent. For Bethe a theory only mattered when he could get actual numbers out.
From Fermi’s Rome, Bethe returned to a Germany whose scientific establishment was nearing the precipice. In his classroom at the ancient university of Tübingen, where he took an assistant professorship, he saw students wearing swastikas on arm bands. It was the autumn of 1932. That winter Hitler took power. In February the Reichstag burned. By spring the first of the Nazis’ anti-Jewish ordinances entailed the immediate dismissal of one-fourth of the country’s university physicists—non-Aryan civil servants. Bethe, his father a Prussian Protestant, did not consider himself a Jew, but because his mother was Jewish his status in Nazi Germany was clear. He was immediately shed from the faculty he had just entered. Across Europe the greatest intellectual migration in history was already beginning, and Bethe had little choice but to join it. Scientists in general had the advantage of working in a polyglot community, where international study and temporary overseas lectureships eased their emotional transition—from citizen to refugee. He reached the New World in 1935.
Feynman had known Bethe’s name since he was an undergraduate—the Bethe Bible, the three famous review articles on nuclear physics, had provided the entire content of MIT’s course. He had seen Bethe once from a distance at a scientific meeting. An ugly man, he had thought at first glance, awkward, with slightly squashed features on a strong frame, light brown hair bristling skyward above a broad brow. Feynman’s first impression dissolved when they met up close in Santa Fe before heading up to Los Alamos for the first time. Bethe, thirty-seven years old, had the body of a mountain climber, and he spent as much time as possible hiking in the canyons or up to the peaks behind the laboratory. He radiated solidity and warmth. Soon after their arrival on the mesa, a statistical fluctuation in the comings and goings of the theorists left Bethe stripped of the people he needed to consult. Victor Weisskopf, his deputy, was away. Teller was away—but Teller, anyway, had immediately grown more aloof than useful; not only had Oppenheimer passed him over in favor of Bethe, but Bethe had passed him over in favor of Weisskopf. So Bethe drifted into Feynman’s office one day, and soon people down the corridor could hear his booming laugh.
Bethe left the initial lectures trying to work out a way of calculating the efficiency of a nuclear explosion. Serber had presented a formula for the simplest case, when the mass of uranium or plutonium was just above critical. For bombs, which would require masses substantially over critical, the problem was far more difficult. He and Feynman developed a method of classic elegance that became known as the Bethe-Feynman formula. The dangerous practicalities of nuclear physics brought other questions. A lump of uranium or plutonium, even smaller than critical mass, raised the possibility of a runaway chain reaction—predetonation. Chemical explosives were far more stable. Bethe assigned this problem to Feynman in the project’s first months. Stray neutrons were always a presence, at some low level of probability—from cosmic rays, from spontaneous individual fissions, and from nuclear reactions caused by impurities. Cosmic rays alone sparked enough fission to make uranium 235 noticeably hotter in the high altitudes of Los Alamos than in sea-level laboratories. Without understanding predetonation, the scientists could not understand detonation itself, because they would not know how the bomb would behave during the split-second transition from subcritical to supercritical. Feynman spent a long time thinking about the properties of a chunk of matter in the peculiar condition of near-criticality, a form of matter that science had not had occasion to ponder before. He recognized that the essence of the problem was not its average behavior but its fluctuations: bursts of neutron activity here and there, spreading in chains before dying out.
Mathematics, in the form of probability theory, had barely begun to provide tools for handling such complex patterns; he discussed the problem with the Polish mathematician Stanislaw Ulam, and Ulam’s approach to it helped midwife a new field of probability called branching-processes theory. Feynman himself worked out a theory of fluctuations building upward from the easier-to-calculate probabilities of short chain reactions: a neutron splits one atom; a newly liberated neutron finds another target; but then the chain breaks. Some measurable fluctuations—audible bursts of noise on a Geiger counter—could be traced back to an origin in a single fission event. Others were combinations of chains. As with so many other problems, Feynman took a geometrical approach, considering the probability that a burst in a certain unit volume would lead to a burst in another unit volume at a given time later. He arrived at a practical method that reliably computed the chances of any premature reaction taking hold. It was suitable even for the odd-shaped segments of uranium that would be blasted into one another in the Hiroshima bomb.
Bethe found in Feynman the perfect foil and goad. This young man was quick, fearless, and ambitious. He was not satisfied to take away one problem and work on it; he wanted to work on everything at once. Bethe decided to make him a group leader, a position otherwise reserved for prominent physicists like Teller, Weisskopf, Serber, and the head of the British contingent at Los Alamos, Rudolf Peierls. For his part Feynman, who had lived through twenty-five years and a full formal education without ever falling under the spell of a mentor, began to love Hans Bethe.
Diffusion
Feynman did some recruiting for the project. He had invited one of his MIT fraternity friends to join the secret work. He even tried to recruit his father. Melville’s health had turned poor—his chronic high blood pressure affected him more and more—and Lucille wished he could afford to travel less. Richard wrote his mother that there might be a job available as a purchasing clerk. He wished, too, that Melville could see at close range the heady intellectual world toward which he had so long aimed his son. “He would be partly out of the rush, etc. of the business work, & would be among academic men to a great extent, which I’m sure he would enjoy … Purchasing these days is quite difficult, & everyone here is in a hell of a hurry for their stuff … it will be a damn important position in our project and scientific venture.”
Nothing came of that suggestion. In the spring of 1944 Feynman came across a familiar name on a list of available physicists: T. A. Welton. He filled out a requisition. His college friend, working as an instructor at the University of Illinois, had been trying to remain a civilian by teaching military-related courses and had watched unhappily as the more distinguished members of his department disappeared to mysterious locales. Feynman’s requisition rescued him. Welton, like so many physicists by then, had pieced together more than the army security officials liked to think possible. When he was invited to meet a stranger in a hotel room in Chicago, and then invited by the stranger to drop everything and move to New Mexico, he understood that this was, as he said later, the classic impossible-to-refuse offer. The day he arrived, Feynman took him on a long hike down into a gorge that had lately been named Omega Canyon. He was able to startle Feynman with an affirmative answer to his first question, “Do you know what we’re doing here?”
“Yes,” Welton said. “You’re making an atomic bomb.” Feynman recovered quickly. “Well,” he asked, “did you know we’re going to make it with a new element?” His friend admitted that the news of plutonium had not drifted as far as Illinois. While they walked—Welton’s lungs desperately drawing in the underpressurized air of 7,000 feet above sea level—Feynman intoxicated him with a briefing. They talked about the bomb. There were now two designs. A uranium bomb would take the form of a gun, creating a critical mass by firing a uranium bullet at a uranium target. A plutonium bomb would use another audacious method. A hollow sphere would be blown inward on itself by the shock from explosives packed all around it. The hot plutonium atoms would be compressed not through one dimension, as in the gun, but through three dimensions. The implosion method, as it was accurately named, was starting to look better and better—in part because so many problems had plagued the alternatives. (Feynman did not mention his own initial reaction when implosion’s inventor, S
eth Neddermeyer, first reported experiments on explosives wrapped around steel pipes. He had raised his hand in the back row and announced, “It stinks.”)
As Welton listened, trying to keep up along the narrow canyon walls, he understood that Feynman was also saying that he had worked hard to establish himself as a smart kid to be reckoned with—that a young researcher had to impress the senior people with his usefulness, that he, Feynman, had been through that process, and that he had succeeded. They talked only briefly about Arline. She was not well, spending most of her days in a wooden bed in the Presbyterian Sanatorium, a small, poorly staffed facility by the side of a highway in Albuquerque. Feynman, visiting her almost every weekend, hitchhiked or borrowed a car to head down the unpaved road toward Santa Fe on Friday afternoon or Saturday. Away from the laboratory he would turn his thoughts back to the pure theory of quantum mechanics. He used the long trip, and the hours when Arline slept, to push his thesis work further. Welton remembered how obstinately his friend had resisted the Lagrangian simplification of dynamical problems when they were a pair of precocious sophomores in MIT’s theory course. He was amused and impressed to hear how far Feynman had taken the Lagrangian method in reformulating the most basic quantum mechanics. Feynman sketched out his idea of expressing quantum behavior as a sum of all the possible space-time trajectories a particle could take, and he told Welton frankly that he did not know how to apply it. He had a wonderful recipe that had not gelled.