YEARS LATER, IN WHAT SHE DESCRIBED AS A “FIT OF HOUSE-CLEANING enthusiasm,” combing through a filing cabinet that stored family papers, Laura Fermi found a copy of the letter Pegram sent to Admiral Stanford Hooper at the Office of Chief of Naval Operations. Dated March 16, 1939, it begins:
Dear Sir:
This morning I had a telephone conversation with Mr. Compton in the office of the Assistant Secretary of the Navy [Charles Edison, Pegram’s friend and son of the inventor], who has doubtless reported the conversation to you. It had to do with the possibility that experiments in the physics laboratories of Columbia University reveal that conditions may be found under which the element uranium may be able to liberate its large excess of atomic energy and that this might mean the possibility that uranium might be used as an explosive that would liberate a million times as much energy per pound as any known explosive. My own feeling is that the probabilities are against this, but my colleagues and I think that the bare possibility should not be disregarded and I therefore telephoned to Mr. Edison’s office this morning chiefly to arrange a channel through which the results of our experiments might, if the occasion should arise, be transmitted to the proper authorities in the United States Navy.
Professor Enrico Fermi who, together with Szilard, Dr. Zinn, Mr. Anderson and others, has been working on this problem in our laboratories, went to Washington this afternoon to lecture before the Philosophical Society in Washington this evening and will be in Washington tomorrow. He will telephone your office and if you wish to see him will be glad to tell you more definitely what the state of knowledge on this subject is at present.
Professor Fermi, formerly of Rome, is Professor of Physics at Columbia University. In December last he was awarded the Nobel Prize in Physics in 1938 for the work he did on the artificial creation of radioactive elements by means of neutrons. There is no man more competent in this field of nuclear physics than Professor Fermi.
Professor Fermi has recently arrived to stay permanently in this country and will become an American citizen in due course. He is very much at home in this country, having visited here often to lecture at the University of Michigan, Stanford University and at Columbia.
Professor Fermi will be staying tomorrow with Professor Edward Teller of George Washington University.
Sincerely yours,
George E. Pegram
Professor of Physics
Across Fermi’s copy of it, Pegram scrawled a note. “Dear Fermi—This may prepare the way for you a little better than Mr. Compton’s explanation to Adm. Hooper.”
Laura was mystified by the letter, having never seen it before and blissfully unaware that her husband met with the Navy on the subject of a potential atomic bomb in March 1939. In what was to become a habit, he had not spoken to her of his initial research relating to the potential for a nuclear explosive. When she confronted him about it, he explained that he had kept it as a kind of insurance policy. In December 1941, when the Axis declared war on the United States, he felt he might need some proof of loyalty to his new country and set it carefully aside in a manila folder. If any authorities challenged him, he would pull it out.
The interchange between Laura and her husband reveals much. Even as early as March 1939, Fermi was maintaining a certain veil of secrecy around his uranium work—this in spite of his disagreements with Szilard over the whole issue of secrecy. The secrecy issue was undecided, but with characteristic caution he decided to say nothing to Laura. The other point, perhaps more telling, is that at the moment the United States entered the war, the Fermis became, albeit for a short time, enemy aliens. Fermi himself was not disposed to take his importance in the war effort for granted and felt the need for documentary proof of his loyalty to the United States. He needn’t have worried. By the time the United States entered the war, he was one of the key players in the effort to build a nuclear weapon and few, if any, doubted his loyalty. At the time, however, he was keenly aware of his status as a foreign national of a potential enemy power and felt the need to have an insurance policy tucked away in his filing cabinet.
Reading the letter from Pegram, Admiral Hooper may have wondered what all the fuss was about. Pegram had to tread carefully. The idea of creating an explosive out of uranium would have struck Navy technical staff at first blush as preposterous, and Pegram believed that overselling the concept would have resulted in doors being shut in Fermi’s face. The letter was designed in part to give the admiral, who may never have heard of Fermi, a sense of the man’s stature within the scientific community. Pegram also went out of his way to assure the admiral that he and his Navy colleagues would be able to understand this foreign national, that Fermi was “very much at home” in America and would make a good presentation to the presumably unworldly Navy staff. In retrospect, tentative though it was, Pegram drafted the perfect letter to initiate contact between the scientific community and the US government on the potential for an atomic bomb.
March 17, 1939, was a cool day in Washington, and the high of forty-nine degrees suggests that the cherry trees around the tidal basin, a 1912 gift from the Japanese people, had not yet blossomed. The basin would have been visible from Navy headquarters, housed in an enormous and exceedingly ugly edifice on what is now beautiful parkland just north of the long, narrow reflecting pool on the west side of Constitution Mall. Fermi arrived to a tepid welcome in the nondescript board room and overheard an unenthusiastic staff member announce to the assembled group, “There’s a wop outside.”
In the conference room, Hooper had assembled a range of technical experts from various offices in the Navy, officers responsible for ordnance, engineering, construction, and repair, as well as a team from the Navy Research Labs headed by Dr. Ross Gunn. Although it was a Navy meeting, a technical team from the Army had been included as a courtesy. Fermi spoke for about an hour providing an overview of the physics of nuclear fission, the potential for the development of nuclear energy, and the prospect of a weapon based on the fission of uranium. According to notes taken by Captain Garrett L. Schuyler, later chief of the Research Division of Ordnance, Fermi described the principles of slow-neutron fission and the importance of neutron emission from fission to create a chain reaction. In summarizing the experiments he and Szilard had just completed, Fermi concluded that “the excess in the number of released neutrons is not very great and has not yet been demonstrated absolutely beyond the possible limits of experimental error.” He added that new experiments were planned in the next months to make more definitive measurements and if “these experiments show more neutrons are released from the uranium atoms than are necessary to split them up, continuous release of energy in a mass of uranium is theoretically possible.” He gave a clear description of critical mass: “In the small samples used so far… the released neutrons are possibly not all effective because some will too rapidly escape; but in a sufficiently large mass of uranium, they necessarily will be all trapped and available in time.”
At this point Captain Schuyler piped up with a question: “What might be the size of this critical mass?” Fermi smiled and gave an answer consistent with his strategy of downplaying the possibility of a nuclear weapon: “Well,” he replied, “it just might turn out to be the size of a small star.” Although Fermi might have been deliberately modulating the practicality of a fission bomb, it is also true that no one—neither Fermi the pessimist nor Szilard the optimist—actually knew what the critical mass might be. Not nearly enough was known to even begin such a calculation.
Fermi turned to the issue of uranium isotopes. He explained to the assembled military scientists that natural uranium consisted of a mixture of the two isotopes, about 99.3 percent of which was uranium 238 (U-238) and 0.7 percent, uranium 235 (U-235). On the basis of theoretical work done the previous month at Princeton by Bohr, Wheeler, and Czech émigré George Placzek, scientists now believed that the far rarer U-235 isotope was responsible for the fission reaction. The only clean way to build a fission weapon would be to separate the two isotopes, but at this point no one knew how to do this.
In summary, Fermi made it clear that it may be possible to unlock atomic energy through fission and that this possibility, with all its attendant uncertainties, should be brought to the attention of the military.
Had he witnessed his colleague’s performance, Szilard would have been disappointed. Fermi was disinclined to stir the Navy into a frenzy of action, and he didn’t. A natural reticence to make bold statements about science before he had clarified the facts for himself, combined with a feeling that the less said about the terrible possibilities the better, drove him to impart enough information for the military to decide on next steps without recommending any specific course of action. Years later, Szilard dismissively suggested that nothing came of the meeting.
This was not entirely true. Fermi’s lecture fired up the Navy’s head of research Ross Gunn, who immediately saw that uranium could provide a source of energy for submarines. He launched a long, frustrating, but ultimately successful effort to develop nuclear-powered naval vessels. Before acting, however, he had to find out more about this odd, understated little man with the strong Italian accent. He called Merle Tuve, the Carnegie Institution physicist who was one of the cosponsors of the January conference at which Bohr and Fermi had presented. “Who is this man Fermi? What kind of man is he? Is he a Fascist or what? What is he?” Tuve assured Gunn of Fermi’s impeccable credentials. That was enough for Gunn. Unfortunately for the Navy scientist, he could not get his own project off the ground until after the war, when national priorities had shifted away from the Manhattan Project.
Though the briefing did not exactly spur the Navy into action, it did result in a check to Columbia for $1,500 for continued research into fission. Who knew? Might there be something to this bizarre, science fiction idea of a new explosive based on nuclear fission? It seemed worthwhile, for a small expenditure, to keep tabs on the research being done at Columbia.
Fermi was sailing in uncharted waters when he arrived on Constitution Mall in March 1939. He came bearing outlandish ideas, ideas that an institutionally conservative military was unprepared to accept. The military was not accustomed to funding private scientific research. In retrospect, it is impressive that he received even a small grant for his experiments.
BACK IN NEW YORK, FERMI BEGAN TO SETTLE INTO LIFE IN HIS NEW country.
The family home at 450 Riverside Drive was one of a row of apartment buildings overlooking the Hudson River, built in the early 1900s to house Columbia faculty. It was reasonably comfortable, but in the winter the wind howled up from the river and up the hill on West 116th Street. Walking the children up the hill or back down against the wind was not one of Laura’s favorite activities. In spite of this, she gradually adjusted to life in a new city, making the best of her situation.
Fermi plunged into classroom work. He taught three courses that spring term, including a course on geophysics, one of his favorite subjects, with a group of rather fortunate undergrads, as well as higher level courses on quantum mechanics and applied quantum mechanics.
He was also getting to know other members of the Columbia faculty, people who would become close colleagues and friends over the next decade. Rabi was one of them. An irascible, punchy personality with a wicked sense of humor, Rabi had met Fermi during his years as a post-doc in Europe. In the early 1930s, Rabi began experimental work that eventually resulted in his discovery of the nuclear magnetic resonance effect that is the basis of today’s MRI scanners. Fermi and Rabi hit it off right away. Years later, Rabi would tell his biographer that aside from Einstein, he considered Fermi the greatest physicist he had ever known.
Another fast friend was Harold Urey. Slightly older than Fermi, Urey won the Nobel Prize in Chemistry in 1934 for his isolation of deuterium, the isotope of hydrogen with a proton and a neutron in its nucleus. Urey befriended Enrico and Laura and spent a good deal of time selling them on the joys of living the American dream in the suburban town of Leonia, New Jersey, where Urey himself lived. Within a year he succeeded, and the Fermis became suburban Americans with front and back lawns and a makeshift workshop in the basement. Fermi never quite got the hang of suburbia—he and Laura were city folk at heart—and their front lawn was often the least well manicured on the block. It was, however, the beginning of a lifelong friendship with the Ureys.
As the family adjusted to life in the States, making new friends and settling into what they hoped would be a quiet domestic life, nuclear fission and Szilard’s obsession with chain reactions continued to preoccupy Fermi.
FERMI HAD TOLD HIS NAVY AUDIENCE THAT HE WAS PLANNING another experiment to clarify some of the uncertainties surrounding fission and the possibility of a chain reaction. For this experiment, Fermi and Szilard collaborated as principal partners. Given their radically different work styles, it is not surprising that this was also their last direct collaboration.
Creating a fission chain reaction using natural uranium taken from the ground poses some real problems. Natural uranium is composed of two isotopes. U-238, which is extremely difficult to split, accounts for 99.3 percent of natural uranium. U-235, which readily splits and is ideal for creating a chain reaction, accounts for only 0.7 percent of natural uranium. Thus, a chain reaction, if feasible, would require many tons of natural uranium, if not separated, to have enough of the fissionable isotope U-235 needed for the reaction. Although no one yet knew how to separate the two isotopes (U-235 and U-238) from each other, Fermi’s Columbia colleague John Dunning doubted the feasibility of a chain reaction based on natural uranium and so pressed for first solving the separation problem, purifying natural uranium so that the relatively small amount of U-235 needed for a chain reaction could be isolated. Doubting that the techniques for isotope separation could be developed quickly, Fermi and Szilard preferred using natural uranium, instinctively sensing that a chain reaction could be developed with a large enough supply of it. After much debate, Fermi’s proposal won out. Fermi recognized that isotope separation would eventually need to be solved but believed that an initial demonstration using natural uranium would provide proof of the chain reaction concept.
Beginning in April 1939, Fermi and Szilard, along with Anderson, modified the water tank experiment, filling the tank with a 10 percent solution of manganese sulfate, a substance that becomes radioactive in proportion to the neutrons that hit it. Into this solution they placed a matrix of fifty-two tin cans, two inches in diameter and two feet high. In the middle of the tank they placed Szilard’s neutron source.
Using four hundred pounds of uranium oxide the ever-resourceful Szilard had “borrowed” in one of his famous sleights of hand from the El Dorado Radium Corporation, which owned significant deposits of uranium ore in Canada, they measured a 10 percent increase in the radioactivity of the manganese with the uranium oxide present inside the cans, confirming the results of the two previous experiments.
Then they set about trying to calculate the ratio of neutrons emitted per neutron absorbed during fission, what they called the “reproduction factor.” A cascade of fission reactions, as Szilard originally envisioned, required an average reproduction rate greater than one—even the slightest amount greater than one would eventually work. The experiment resulted finally in a measurement of about 1.5 fast neutrons emitted for every neutron absorbed during fission. They reported that “a nuclear chain reaction could be maintained in a system in which neutrons are slowed down without much absorption until they reach thermal energies and are then mostly absorbed by uranium rather than by another element.” They also suspected that the water they used in the experiment was absorbing too many neutrons: “It remains an open question, however, whether this holds for a system in which hydrogen is used for slowing down the neutrons.” This was the first conceptual outline of what was to become the “pile,” the world’s first nuclear reactor in 1942.
Anderson later noted several important points about the experiment.
First, the measurements they took indicated that plain water would probably not be a good moderator for a fully functioning reactor, because hydrogen atoms in the water had a tendency to absorb slow neutrons into their nuclei, taking them out of the chain reaction.
Second, the team became aware of the importance of a phenomenon known as “resonance absorption.” U-238, which makes up the vast majority of natural uranium, tends to absorb slow-ish neutrons without undergoing fission, thus taking them out of the chain reaction. Fermi estimated that this phenomenon was responsible for a 20 percent decrease in the average number of emitted neutrons, an estimate based as much on Fermi’s intuition as on the data. To solve this problem, Fermi decided that lumping uranium into smaller chunks would reduce the tendency of the fast neutrons emitted in fission to slow down and become absorbed by the U-238.
Anderson went out of his way to explain why this was the last time that Fermi and Szilard collaborated directly on an experiment:
This was the first, and also the last, experiment in which Szilard and Fermi collaborated together. Szilard’s way of working on an experiment did not appeal to Fermi. Szilard was not willing to do his share of the experimental work, neither in the preparation nor in the conduct of the measurements. He hired an assistant to do what we would have required of him. The assistant, S. E. Krewer, was quite competent, so we could not complain on this score, but the scheme did not conform with Fermi’s idea of how a joint experiment should be carried out, with all the work distributed more or less equally and each willing and able to do whatever fell to his lot. Fermi’s vigor and energy made it possible for him to contribute somewhat more than his share, so that any dragging of feet on the part of the others stood out more sharply in contrast.
Fermi was never so explicit. During his January 1954 lecture to the American Physical Society about this period at Columbia, all he would say about his brilliant but frustrating collaborator was that he was “a very peculiar man, extremely intelligent,” a description that brought hearty laughter from the audience. Whatever reservations Fermi had about Szilard’s willingness to get his hands dirty in the lab—and uranium oxide is very dirty—he retained enormous respect for Szilard’s ability to think creatively about difficult scientific problems.
Fermi, Szilard, and Anderson submitted their experimental results for publication in Physical Review in early July 1939. Soon Fermi and the family were off to Ann Arbor, Michigan, for another summer school session, where he lectured on the absorption of cosmic rays in the atmosphere and in solids. He also met up with an old acquaintance from his days in Germany, and they had a conversation that lingered in Fermi’s mind for the next six years.
IN LATE JULY 1939, WERNER HEISENBERG ARRIVED IN ANN ARBOR to spend a week participating in the Goudsmit-Uhlenbeck summer session, seeing old friends and discussing the state of the world.
The state of the world was grim. To anyone reading the daily reports coming out of Berlin, Moscow, Paris, and London, it was clear that European powers were preparing for an outbreak of hostilities in the very near future. The betting was that Germany would invade its ideological enemy, the Soviet Union, within the month.
At that moment, Heisenberg arrived on the scene in Ann Arbor.
Max Dresden, a young student tending bar at a party hosted by refugee physicist Otto Laporte, witnessed an encounter between Fermi and Heisenberg. Dresden, who would go on to a distinguished career at Stanford, describes the evening:
There was actually not much to do, so we could pay close attention to the conversations. There was really only one central topic. Fermi had just left fascist Italy to come to the US; Heisenberg had decided to return to Nazi Germany. The crucial part of their argument was whether a decent, honest scientist could function and maintain his scientific integrity and personal self respect in a country where all standards of decency and humanity had been suspended. Heisenberg believed that with his prestige, reputation and known loyalty to Germany, he could influence and perhaps even guide the government in more rational channels. Fermi believed no such thing. He kept on saying: “These people [the Fascists] have no principles; they will kill anybody who might be a threat—and they won’t think twice about it. You have only the influence they grant you.” Heisenberg didn’t believe the situation was that bad. I believe it was Laporte who asked what Heisenberg would do in case of a Nazi-Soviet pact. Heisenberg was totally unwilling to entertain that possibility: “No patriotic German would ever consider that option.” The discussion continued for a long time without resolution. Heisenberg felt Germany needed him, that it was his obligation to go back. Fermi did not think there was anything anyone could do in Italy (or Europe); he was afraid for the life of his wife (her father was later killed); and so he felt it was better to make a fresh start in the US. But none of the decisions had come easy. The role of physics and physicists was mentioned off and on. After the party was over everybody left in a state of apprehension and depression.
Some three decades later Heisenberg recalled another one-on-one conversation with Fermi, at Fermi’s Ann Arbor apartment. Fermi started out on a positive note, describing how his move to the United States was liberating, how the United States had been home to European refugees for generations, and how stimulating it was to start all over again in his new homeland. “Here, in a larger and freer country, [Europeans] could live without being weighed down by the heavy ballast of their historical past. In Italy I was a great man; here I am once again a young physicist, and that is incomparably more exciting. Why don’t you cast off all that ballast, too, and start anew?” Heisenberg replied that he understood the attraction, but that abandoning Germany now, he would feel himself a traitor, particularly to younger physicists who did not have the ability to emigrate and to find work wherever they wanted. For Fermi, however, any responsibility he felt for the students he left behind was outweighed by the many compelling reasons he had for leaving Italy. In later years, Fermi’s notably generous treatment of his American students may have been an effort to compensate for lingering feelings of guilt he had over abandoning his Italian students.
Fermi pressed on, explicitly referring to the possibility of using the discovery of atomic fission to create a bomb. He warned that Heisenberg would be expected to work on such a project. Heisenberg expressed doubt that such a weapon could be built, at least not quickly. Fermi then asked, “Don’t you think it possible that Hitler may win the war?” Heisenberg expressed doubt, given the balance of technological resources available on each side. Fermi was incredulous that under the circumstances Heisenberg still wanted to return to Germany. Heisenberg explained that patriotism was a stronger factor for him than doubt about the war’s outcome.
Fermi ended the conversation, noting, “That’s a great pity. Let’s just hope we will meet again after the war.”
Fermi may have had reservations about the possibility and wisdom of pursuing an atomic weapon. He had soft-pedaled the idea in his meeting with the Navy in March 1939 and continued to have doubts about the technological feasibility of the weapon. His encounter with Heisenberg dramatically altered his perspective. He had known Heisenberg since 1922 and, though Fermi might not have particularly liked the man, he had followed his career and his contributions with great interest. He had even nominated Heisenberg for a Nobel Prize. Heisenberg would now return to his home country and when the inevitable war broke out would be tasked by Hitler with developing a weapon based on nuclear fission. Fermi had enough respect for Heisenberg to know the serious threat he posed. Whatever reservations Fermi had about the project, he would now have to pursue it with vigor. He had no real choice in the matter.
THE RESULTS OF THE SPRING EXPERIMENT WITH SZILARD AND Anderson were never far from Fermi’s mind. Over the summer he corresponded with Szilard on a central problem: if water was not a suitable moderator for the chain reaction, was there another substance that would be suitable?
Their thoughts turned to a form of carbon called graphite.