Chain-Reaction “Obsession”
1934–1938
“I feel somewhat depressed about Szilard,” Eugene Wigner wrote to their mutual friend Michael Polanyi in January 1934, because “his last letter sounded a little more unstable than I am used to with him.”1 Wigner and others fretted about finding Szilard a steady job: at Princeton, where Wigner taught physics; at New York University, where Wigner hoped Szilard would “sooner or later feel quite happy.”2
Szilard infuriated his friends, at first telling Polanyi he was “very sorry” when a Liverpool University appointment fell through but in the next sentence asking to visit him at the University of Manchester along with the Berlin physicist Fritz Lange, “more for the fun of it than for any special purpose. . . .”3 Szilard had met Lange and his partner, Arno Brasch, in Berlin during the 1920s and brainstormed with them about accelerators, which use high voltages to speed up radioactive particles to bombard atoms. For maximum voltage, Brasch and Lange were trying to tap lightning at alpine research stations atop the Zugspitze on the German-Austrian border and on Monte Generoso near Lake Como—a project that both amused and frightened Szilard.4
At the time, accelerators seemed a promising way to manufacture radioactive isotopes for use in medicine, and during Lange’s December visit to England, he and Szilard toured the General Electric (U.K.) research laboratory at Wembley, in west London.5 Szilard considered working there, briefly, and he had fancied teaching physics in India, also briefly.6 But despite fervent efforts by Wigner and other friends, Szilard had “decided to mark time” in his career.7
Even more than lightning-powered accelerators, however, it was his nuclear-chain-reaction concept that energized Szilard’s thinking. His selfproclaimed “obsession” with chain reactions distracted Szilard from the refugee-settlement work for the Academic Assistance Council (AAC), and by March 1934 he had reduced his ideas to paper. In a fifteen-page patent application he named beryllium as the element most likely to be split by neutrons and, in turn, free other neutrons in a spontaneous frenzy of energy. In fact, beryllium would never split as Szilard expected; he was misled by incorrect published data about the element’s atomic weight. But, intuitively, Szilard also named uranium and thorium—the only two natural elements that would eventually sustain chain reactions.8 He filed the patent on Monday, March 12, 1934, handing in a typed manuscript sprinkled with penned corrections and x-ed-out words. Then he began to dream about the atom’s new commercial uses, perhaps replacing coal and oil as the world’s industrial fuel; about its social implications, perhaps bringing abundant energy to developing countries now starved for water and minerals; and—unavoidably—about its potential as a weapon of mass destruction, perhaps giving Adolf Hitler “atomic bombs” to terrorize the world.
His patent filed, Szilard wrote to Sir Hugo Hirst, founder of General Electric (U.K), then on holiday at the Carlton Hotel in Cannes, and in a playful letter urged him to read “a few pages” from The World Set Free by H. G. Wells. In the book Szilard cited an “interesting and amusing” section that had predicted in 1914 how artificial radioactivity would be produced in 1933, just as it had been by the Joliot-Curies.
Of course, all this is moonshine, but I have reason to believe that in so far as the industrial applications of the present discoveries in physics are concerned, the forecast of the writers may prove to be more accurate than the forecast of the scientists. The physicists have conclusive arguments as to why we cannot create at present new sources of energy for industrial purposes; I am not so sure whether they do not miss the point. . . .9
But instead of proving this point, Szilard pursued another invention the same week and filed a patent for a “microbook” that would reduce whole libraries to minute images on a roll of film. He would later learn that Siemens had already patented the idea, as “microfilm.”10
Also in mid-March, Szilard’s mentor and friend Albert Einstein wrote to support a Rockefeller Foundation grant that would finance Szilard’s research at NYU, praising him as “an especially intelligent and manysided scientist, extraordinarily rich in ideas.” Einstein mentioned Szilard’s “absolutely independent performance” in his Ph.D. thesis and his original paper on entropy and information, noting also that “he is the originator of an important feature of the modern quantum mechanics, namely, of the consideration of measurements in the formulation of the wave function.” Szilard’s “personality is estimated very highly by all his colleagues,” Einstein wrote, “especially for his great unselfishness.”11 While praising Szilard’s originality, Einstein would not learn about the truly original chain-reaction concept for another five years.
With no laboratory of his own to test his chain-reaction ideas, Szilard enlisted Fritz Lange and Lise Meitner in Berlin to arrange certain experiments into “the production of radioactive bodies.” Szilard urged them to “take one after another all seventy elements and bombard them with cathode rays [X-rays] and see if there is any activity by using a Geiger counter or the Wilson [cloud] chamber.”12 In science at the time, such international collaboration was rare, although to Szilard it seemed the obvious thing to do.13
For all his involvement with nuclear research, Szilard could not ignore the pull of politics. It caught him when he picked up the Manchester Guardian on the morning of April 24 and read that Japan had rejected all interference, by the League of Nations or by any country, in its invasion of Manchuria. This invasion, and Japan’s arrogance, upset Szilard’s “sense of proportion,” the moral and ethical balance that he sought in nature and in modern life. Szilard ripped the article from the page and with a letter mailed it to Lady Murray, the wife of classical scholar Gilbert Murray, then president of the International Committee of Intellectual Cooperation and chairman of the League of Nations Union. “We felt [at NYU in 1932] that a mere protest would not be of any value,” Szilard wrote, “but that a definite pledge on the part of the leading scientists, though of rather limited value in itself, would serve to ‘keep the faith’ in the cause of justice.”
In his current scheme, Szilard would try to politicize Nobel laureates for the first time. But this needed “the right person” to call for a scientists’ boycott of Japan. Realizing how timid his scientific colleagues could be, Szilard added a twist to his appeal: The laureates’ protest would only take effect if “eight-tenths” of all prize winners agreed to sign. Under his proposed boycott, scholars would refuse to send scientific and technical information and journals to Japan or to cooperate with Japanese students. 14
Gilbert Murray liked Szilard’s idea, “considering the extreme reluctance of governments to shoulder any burden or take any risk for the sake of world peace,” and urged him to call on Dr. Maxwell Garnett, at the League of Nations Union, to consider approaching some famous scientist for the cause. But when Szilard met Garnett and realized that the plan appeared “too rigid and impractical” to him, he let his efforts lapse, although enlisting Nobel laureates in political causes would eventually become a popular technique.15
Wigner and Szilard met in London early in the spring of 1934, and the two talked for hours about the chain-reaction patent. From Wigner, Szilard obtained new calculations and—most important to him—encouragement from a brilliant colleague whom he respected. Szilard reduced the chain-reaction idea to a simple equation: Be9 + n = Be8 + 2n. Bombard beryllium (Be) with a neutron (n). The beryllium changes atomic weight, but not its chemical identity, and in the process emits two neutrons. These neutrons penetrate other atomic nuclei, releasing four neutrons, then eight, then sixteen, and so on. Control this instantaneous reaction and you create immense quantities of heat; let it run wild and you transform matter to energy in a violent explosion.16
In May, Szilard warned Australian physicist Mark Oliphant about a possible nuclear explosion, but he seemed to make little of it.17 During the last week of that month, Szilard and Wigrer visited the Cavendish Laboratory in Cambridge, where the director, Nobel laureate Ernest Rutherford, had agreed to meet Szilard “for a short time,” although
offering “little chance of . . . getting work” there.18
Their talk, at noon on Monday, June 4, 1934, might have changed the course of nuclear research and with it modern history. Szilard, the author of the nuclear-chain-reaction concept, met Rutherford, the authority who could have tested and developed that idea. But Szilard was overly cautious, perhaps nervous, and in the great man’s presence apparently tried to appeal to Rutherford’s scientific conservatism. Szilard described a chain-reaction method that was more conventional than the neutron theory that had flashed before him in Southampton Row. Szilard said that a chain reaction might be achieved in two stages: Alpha particles would bombard light elements to release their protons, as in the many transmutation experiments performed at the Cavendish Laboratory by Rutherford and Chadwick. These protons, in turn, would disintegrate lithium and release two alpha particles, as Cockcroft and Walton had done there when they first split atoms.
In this explanation, Szilard kept secret his original idea that two particles would escape with each bombardment. And instead of neutrons, he used Rutherford’s technique with alpha particles. It was a silly mistake. Rutherford knew the atomic structures and energy efficiencies of the systems Szilard described. Szilard did not. This apparent ignorance exasperated Rutherford, who became further upset when Szilard announced that he had filed a patent on the chain-reaction concept. When Szilard asked if he could work at the Cavendish Laboratory to test his ideas and offered to provide recommendations, that did it. “I was thrown out of Rutherford’s office,” Szilard later told his friend Edward Teller.19
Three days after this testy meeting, Rutherford’s answer arrived by mail at the flat of Szilard’s friend Esther Simpson in north London, and fearing it contained a rejection, Szilard sat down to write Rutherford, On one of the few occasions in his life when he groveled, Szilard wrote that “to be able to work in the Cavendish Laboratory for a year means so very much to me that I feel extremely anxious that a decision about the possibility of my entering there should not be based entirely on the few conversations I have recently had in Cambridge.” He promised to arrange recommendations from von Laue and Schrödinger. He promised, as Rutherford had suggested, to meet with a Professor Fowler in Cambridge. But the “unusual request” that followed must have enraged Rutherford even more than their face-to-face meeting. “I have just heard that a letter from Cambridge has arrived at my Halliwick Road address,” Szilard wrote, “and I would ask you, should that letter be from yourself containing your decision, to allow me to return it to you unopened.”20 There is no record in the Szilard or the Rutherford papers that Rutherford ever replied or that Szilard ever pressed his request.
To financiers, Szilard appeared just as impertinent and impractical. He asked investors to fund experiments into “a rather romantic enterprise on which I embarked,” one “which arose out of certain recent developments in physics.” These experiments, Szilard promised, “could in a short time lead to a sort of industrial revolution,” although “it is not possible to foretell with certainty the outcome of the proposed experiments.”21
Throughout the spring and summer of 1934, Szilard repeatedly approached GE officials with offers they politely termed “somewhat vague.” These letters document Szilard’s inability to communicate his mind’s intellectual excitements. He could scarcely focus on the chain-reaction concept himself, and during the six months he sought GE’s help, he filed as many “improvements” to his original patent: adding the names of elements likely to release neutrons, giving the size of the beryllium block to be used in experiments, proposing to mix seventy elements together in order to isolate the radioactive ones systematically. While promising to tell the GE executives about the atom’s new “industrial applications,” Szilard would only describe his work on medical isotopes. The chain-reaction experiments, which Szilard called “the other more important issues,” he kept secret, proposing instead to “give a detailed picture to some third person who is attached to one of the English universities and that you should get information from him about his views on the subject,”22
Understandably, GE researchers could see nothing “new” or “practical” in the techniques to manufacture isotopes and asked, instead, for “concrete proposals” on Szilard’s mysterious power source—provided his proposals are contained in patents.23 This request prompted Szilard’s “Memorandum of Possible Industrial Applications Arising Out of a New Branch of Physics,” which predicted “liberating atomic energy” on “such a large scale and probably with so little cost that a sort of industrial revolution could be expected; it appears doubtful, for instance, whether coal mining or oil production could survive after a couple of years.” Szilard mentioned holding patents in the field but would only cite those for isotope production, not for the chain reaction. No wonder GE and other potential investors were skeptical.
Had Szilard been more practical and disciplined, his chain-reaction concept might have been confirmed first in the Bronx, at NYU’s physics department, which in the summer of 1934 had invited him to work for a year as a research associate. At the time, Szilard’s “Suggested Experiments for the Detection of Nuclear Chain Reactions and the Liberation of Nuclear Energy” were in a memo that proposed both commercial and university work. But Szilard was unsure if NYU’s laboratories were properly equipped, and he miffed physicists there by demanding the right “to resign at the beginning of the term if the equipment disappointed him.”
24 Szilard also annoyed GE executives, who told him that his accelerator work was not original and that his “larger issue” of power production— the chain reaction—”is so far outside the scope of a company’s normal activities, that unless the proposition takes some much more definite shape, it would be impossible to participate.”
“I am afraid I have to contradict almost every statement you make in your last letter,” Szilard countered in his impolitic style. GE’s research director eventually apologized for his “inadvertent error,” but that ended the company’s interest in Szilard’s revolutionary new energy source.25
Seeking an academic laboratory for his chain-reaction experiments, Szilard called on University of London physicist George Paget Thomson and dazed him with “as many details as is possible in one interview.”26 Thomson, who would be a Nobel laureate in three years, appreciated Szilard’s “quite interesting ideas” and offered him a place in the laboratory. The “commercial possibilities” of Szilard’s ideas held “a fairly remote chance of a very important discovery,” Thomson concluded, and he thought “that the chance of atomic disintegration becoming of commercial importance in the future is very real, and the type of experiment which Dr. Szilard proposes seems as likely to lead to it as any other.”27
Despite this enthusiastic response, Szilard never pursued Thomson’s offer, perhaps because the results would have to be published in open scientific journals and might alert the Germans to the chain reaction’s possibilities. Instead, Szilard wrote to Francis Simon in Oxford, a German physicist he had known in Berlin, seeking through him a meeting with Frederick A. Lindemann, the physicist who directed Oxford’s Clarendon Laboratory and who advised Imperial Chemical Industries (ICI) about its research fellowships. Lindemann had studied with Walther Nernst in Berlin and during the world war had invented a way to save an aircraft from an uncontrolled spin. He had also developed a formula for determining different materials’ melting points.28
But before his university prospects in New York, London, or Oxford developed, Szilard had talked his way into another research laboratory. Interest in manufacturing isotopes for medical use led Szilard to St. Bartholomew’s Hospital, a teaching institution near St. Paul’s Cathedral. Szilard had met the director of the physics department, F. L. Hopwood, and in early August 1934 gained his permission to use radium samples for research during the summer holidays, provided Szilard team up with a hospital staff member.29 Soon Szilard and Thomas A. Chalmers, a young research physicist, were studying beryllium—the element Szilard believed would produce a
nuclear chain reaction.
At the time, Szilard had stopped working for the AAC but kept in touch with his colleague Esther Simpson. She had moved from north London to a flat on Brunswick Square in Bloomsbury. Through August and September Szilard divided his time between the hospital, where he worked longer and longer hours; his hotel, where he ate and slept and soaked in the bathtub; and Miss Simpson’s, where he dropped by for tea and chitchat and also received mail.
To test whether gamma rays from radium would free neutrons from beryllium, Szilard and Chalmers surrounded an ounce of radium with a thin sheet of beryllium, then wrapped this with silver, iodine, or iridium.30 They set a Geiger counter on the outer metal sheet to detect neutrons and during their collaboration made two important discoveries. First, they learned that beryllium did emit neutrons when exposed to the radium’s gamma rays. Nothing like a chain reaction followed, but Szilard now saw for himself that neutrons could act to rearrange and transform (transmute) the structure of certain atoms.
In Rome a few months earlier, physicist Enrico Fermi and his colleagues had begun to bombard different elements with neutrons. They found that slow neutrons, with very low energy levels, were more easily absorbed by an atom’s nucleus than were fast (higher-energy) neutrons. Once its nucleus absorbed a neutron, the atom became a different “isotope” of the element—having the same number of protons but differing numbers of neutrons and thus chemically related to its progenitor. This new arrangement within the atom usually made its structure unstable and resulted in a release of energy as radiation.
Sometimes Fermi’s group noted that a neutron added to an atom’s nucleus did not simply make it a heavier isotope of itself. Instead, it became an isotope of a different element on the periodic table. For example, when rhodium103 absorbed a neutron it did not become rhodium104; it became palladium104. In this way indium115 became tin116. This phenomenon of an atom absorbing a neutron and becoming a heavier element came to be called the Fermi effect.
Genius in the Shadows Page 21