The Headmaster and His School

The Laboratory's scientific and technical staff arranged within and on top of the magnet of the 60-inch cyclotron, 1939.

The Federal Telegraph magnet did not enter the old Engineering Testing Laboratory, renamed the "Radiation Laboratory," until January 1932. While it was being outfitted at the Pelton Water Wheel Company in San Francisco, Livingston and David Sloan, whom Lawrence had found at the General Electric Research Laboratory and persuaded to come to Berkeley as a graduate student, improved cyclotron technique. Livingston built an 11-inch cyclotron and installed it in Room 329 LeConte Hall. Together with Sloan, who was tuned to the state of the art in industrial electronics, he built a high-power radio frequency oscillator with a Federal Telegraph water cooled tube. It gave 50 kilovolts of accelerating potential at frequencies up to 20,000,000 Hertz. About the time the great magnet was moving into the new laboratory, the 11-inch cyclotron in LeConte gave out one billionth of an ampere of 1.22 MeV protons. "Lawrence literally danced around the room with glee," Livingston recalled. "With 20,000,000 [eV]," a friend calculated, pretty closely, "you'll get the Nobel prize."

Sloan's main job for Lawrence in 1930 and 1931 was building a bigger version of Wideroe's little linear accelerator of heavy ions. In 1931 Sloan completed a tube 1.14 meters long with thirty electrodes that yielded almost one millionth of an ampere of 1.2 MeV mercury ions. Lawrence's growing armament could shoot million volt particles from either end of the periodic table. With neither the linac nor the cyclotron, however, did Lawrence's associates do much nuclear physics. Sloan was reassigned to a project designed to keep alive philanthropic interest in the Rad Lab. Lawrence's backers, the Research Corporation and the Chemical Foundation, had just succeeded in breaking General Electric's patents on high-energy x-ray tubes. At their request Sloan developed a competitor, later installed at the University of California Hospital in San Francisco and the Crocker Institute for Cancer Research at Columbia. In hopes of supporting other scientific enquiries by its investments in accelerator technology, the Research Corporation patented not only the Sloan x-ray tube, but also the cyclotron and the Van de Graaff accelerator.

The 11-inch cyclotron, shown installed in Room 329 Le Conte Hall, was the first cyclotron to exceed 1 MeV.

The making of million-volt protons in January 1932 appropriately opened a year of exceptional discoveries in nuclear science. The same month Harold Urey and his collaborators at Columbia declared the existence of a hydrogen isotope twice as heavy as the ordinary kind. In February James Chadwick announced his discovery of the neutron at the Cavendish Laboratory in Cambridge, England. In April John Cockcroft and Ernest Walton, also at the Cavendish, succeeded in disintegrating lithium atoms with 125 kV protons from their voltage multiplier. In the fall Caltech's Anderson found the positron. And throughout the year Lawrence, Livingston and Sloan labored to produce a beam between the poles of their 75 ton magnet. The sheet metal tanks that held the cooling oil leaked. "We all wore paper hats," Livingston recalls, "to keep the oil out of our hair." Experimentation with shimming gradually brought the beam to larger radii and energies; two symmetric dees were installed; and in December the new 27-inch cyclotron produced 4.8 MeV hydrogen ions.

The artificial disintegration of nuclei was one of the purposes of the apparatus Lawrence had designed. The disintegration of lithium might have been achieved at Berkeley before its accomplishment at Cambridge, and perhaps more easily because of the eight-times greater energy of the California protons. But the planning of physics experiments had not paralleled the construction of the instruments to perform them. This negative consequence of Lawrence's concentration on accelerator improvement was to recur throughout the 1930s. Artificial radioactivity and nuclear fission, to mention only the most dramatic cases, could well have been found at Berkeley; they were certainly produced there before being noticed elsewhere. In the case of artificial disintegration, the Laboratory lacked the proper detectors. Lawrence asked his old friend, Donald Cooksey of Yale, a masterly instrument maker, to provide what was needed. Cooksey and a student of his, Franz Kurie, built the detectors at Berkeley during the summer of 1932. They allowed the Rad Lab to confirm and extend the transformation first accomplished at the Cavendish.

The discovery of deuterium (as Urey called heavy hydrogen) also had strong consequences for Lawrence's program. In March 1933 his colleague in chemistry, G. N. Lewis, who had the largest reservoir of heavy water in the world, gave Lawrence enough to use as projectiles for the developing 27-inch cyclotron. For a time he had a quasi-monopoly of fast deuterons, which, he hoped, would help bring to Berkeley the lead in nuclear physics that the Cavendish then enjoyed. The performance of the deuteron exceeded his most extravagant expectations: it appeared capable of disintegrating every nucleus heavier than helium. But the higher Coulomb barrier presented to the deuteron by the heavier elements made this hypothesis unlikely, and Lawrence, Lewis, and Livingston claimed instead that on collision with just about anything the deuteron itself splits into its constituent proton and neutron. An argument with other nuclear laboratories ensued. It turned out that Lawrence's group had dirtied the cyclotron with deuterium, and that their fast protons arose from the interaction of the deuteron beam with the heavy-hydrogen contaminant. (See Episode beginning page 18.)

Early in 1934 Frederic Joliot and Irene Joliot-Curie, working at the Institut du Radium in Paris, made the discovery that brought them the Nobel prize and redirected much of experimental nuclear physics. In investigating the emission of positrons from aluminum struck by alpha particles, they observed that the target stayed active after the bombardment stopped. It was a great surprise. Everyone had tacitly assumed that the explosion of a nucleus followed immediately on its swallowing an energetic particle, and had arranged his experimental practice to suit. At the Rad Lab belief that residual activity does not exist affected operations in at least two ways. First, no one thought about protection against radiation when the cyclotron was not running (and little enough when it was). Second, the detecting instruments and counters were not set to register electrons and gamma rays. Does the falling tree make a noise if no ear hears it? The cyclotron had been producing substances with much stronger artificial radioactivity than the little bit of radioactive phosphorus the Joliots found, but no detector had listened. Lawrence and his students reproduced the French discovery within a half hour after reading about it in Nature. A weekend's work bombarding twelve elements with deuterons produced as many new activities. The subject was unexpectedly rich. "We are rather bewildered," Lawrence wrote his old friend Jesse Beams. "Already it is clear that nuclear physics offers a very extensive and complicated and interesting field of investigation.

Then Enrico Fermi's group in Rome showed that neutrons induced activity in practically all the elements. Lawrence, who had advertised possession of the world's most powerful neutron beam (formed by irradiating beryllium-9 with ten billionths of an ampere of accelerated deuterons) once again confirmed and extended European results, and expressed surprise at the richness of nuclear transactions. From March of 1934 until the Laboratory went to war, the investigation and production of artificial isotopes by neutron, proton, deuteron, and alpha-particle beams dominated its research program.

Lawrence committed the Laboratory to this program for several reasons. First, the detection and identification of new activities gave information about nuclear reactions and systematics and helped to determine conditions of stability. Nuclear scientists at Berkeley mapped the limits of the isotopic range of the known elements and, in 1940, pushed beyond uranium. Second, Lawrence knew from the work of Georg von Hevesy that radioactive tracers in the body could give unique information about metabolism and other physiological processes. The cyclotron could not only produce tracers in larger amounts than easily available in nature, it might also, and more importantly, create new radioisotopes with properties particularly adapted to biological research. No doubt the expressed interest of the Rockefeller Foundation and the Macy Foundation in the application of the techniques of the physical sciences to the life sciences encouraged Lawrence's attention to the creation of material for biomedical research. He received substantial sums from both philanthropies, and in turn supplied established workers like Hevesy with biologically active radioisotopes. The easy availability of these substances at Berkeley interested local faculty in tracer work, culminating in the discovery of carbon-14 by Martin Kamen and Samuel Ruben in 1940.

Soon after he began his search for useful radioisotopes, Lawrence had the good luck to make sodium-24 efficiently by bombarding rock salt with deuterons. The new substance runs through the body like ordinary sodium; its convenient half-life, fifteen hours, made it useful in diagnosis and therapy. "My medical friends tell me that the properties of radiosodium are almost ideal for many medical applications, such as the treatment of cancer." Lawrence predicted that sodium-24 would supersede radium, and to make sure he promoted it on a national lecture tour. A volunteer--the first two were Alvarez and Joseph Hamilton of the University's hospital in San Francisco--would down a solution of the isotope, and Lawrence would track its course through his body. Audiences appreciated this up-to-date natural magic with material less disagreeable, though no easier to procure, than skull moss or unicorn's horn. Lawrence received fresh supplies of sodium-24 by air mail just in time for these lectures, which increased the drama, and the value, of radioisotopes.

Joseph Hamilton drinking radiosodium, 1939; at right is R. Marshak.

Radiosodium did not fulfill Lawrence's hopes. Other isotopes generated by his cyclotron, however, found important applications in medicine. Phosphorus-32 has been used successfully in the treatment of leukemia, polycythemia vera, other bone-marrow disorders, and Hodgkins disease; iodine-131 in the treatment of thyroid disease; and cobalt-60 in cancer chemotherapy. Perhaps the most interesting of these substances to the physicist and chemist is technetium-99, used in cancer diagnosis. Technetium, element 43, which occupies one of the four places in the periodic table still vacant in 1935, does not exist naturally. It was found in a molybdenum deflector strip from the 27-inch cyclotron, where quantities sufficient for radiochemical analysis had accumulated during months of exposure to fast deuterons. Lawrence presented this object to Emilio Segre, who visited the Laboratory in the summer of 1936 and took the "invaluable gift" to Italy, to stimulate nuclear science at the University of Palermo, where he had recently become a professor. In June 1937 Segre's group announced the first element made by man. Medical application of the new element began in 1947. Half of the seventy artificial radionuclides in common use in medicine today first made their appearances in cyclotrons, and half of these were discovered, or first synthesized, at the Radiation Laboratory.

The potential of radioisotopes for biological research and medicine gave a third reason for the search for new radionuclides: support of further cyclotron development by the sale of active material. Lawrence planned to reap the benefits indirectly, through grants from the Research Corporation, to whom he suggested patenting his method of making sodium-24. The Corporation did not succeed in obtaining a patent on radioisotope production by deuteron bombardment, but their patent on the cyclotron probably would have protected commercial production of radioisotopes until the invention of a different and more prolific source, the nuclear reactor, during the second world war. Although a radiopharmaceutical industry did not materialize in the 1930s, the hope that it might helped to sustain accelerator physics.

In 1936 the University of California officially established the Radiation Laboratory as an independent entity within the Physics Department. This reorganization brought a post of assistant director, to which Lawrence named the indispensable Cooksey, who had been living on his private income; provision for research students; and a promise of help in raising money for the next cyclotron. The generosity of the University, mediated by Sproul, had been stimulated by an attempt by Harvard to acquire Lawrence as Dean of its school of Engineering.

The reorganized laboratory was dedicated to nuclear science rather than, as in its first incarnation, to accelerator physics. This transformation, as we know, resulted from opportunities opened by the discoveries of artificial radioactivity and the biological action of neutron rays, and also, perhaps, from concern about the effects of the increasingly intensive neutron background on the men who worked around the accelerator. A center for nuclear medicine already existed at the University of California Hospital in San Francisco, where Hamilton and Robert Stone operated the x-ray tube built by Sloan. They were joined by Lawrence's brother John, who had been interested in the biological effects of neutrons during a visit to Berkeley in the summer of 1935. Money for the machine promised Lawrence in 1936 was raised on the ground of its utility in medicine. The Chemical Foundation pledged $68,000 for a "medical cyclotron," which was to be the special instrument of John Lawrence and his associates. University Regent W. H. Crocker, who had provided for the Sloan tube, gave what was needed to house the new machine; and by the time the 60- inch cyclotron first operated in June, 1939, spitting 16 MeV deuterons a meter and a half through the air, the Rockefeller Foundation and the National Advisory Cancer Council had also contributed.

Glenn Seaborg, who came from G. N. Lewis's College of Chemistry, and Kamen, from the University of Chicago, put the chemistry in nuclear chemistry in 1937. With J. J. Livingood, Philip Abelson, Segre, and McMillan, they determined the nuclear and chemical characteristics of a great many new substances. Kamen soon had responsibility for making the isotopes required in biological investigations. Everyone added to the list, which, by 1940, amounted to about 40 percent of the 335 artificial radioactive isotopes then known.

A prewar cartoon of the 60-inch cyclotron.

Accelerator design and improvement also changed in scope in 1936. Until then the staff, mostly physicists, had improvised as they went along, shimming here and there, servicing only at breakdowns (which were often enough), and struggling to maintain a vacuum in the cyclotron tank against the 30 ton forces on its iron lid. Because Lawrence had distrusted rubber gaskets, the vacuum was made tight with sealing wax, which tended to crack at places subject to large mechanical stresses. These recurrent faults, named after their discoverers--"Henry's hole," "Luie's leak," "Art's orifice"--would be inspected when the tank pressure became too high for operation. But often enough hours and sometimes days went into the search. The staff needed its well-known optimism and camaraderie to put up with the sulks and tantrums of the early machines.

The centerpiece of the Lab was the expanded 27-inch cyclotron. Late in 1936 Lawrence began to enlarge the pole pieces, as he had hoped to do since 1932. Even before the 27-inch was running he thought the yoke could support pole pieces almost 65 percent larger than those planned, but his budget could not stand the additional cost. Also, larger pole pieces and the consequent higher energy required expenditures for cooling systems for the vacuum structure and magnet. In the more comfortable circumstances of the Laboratory in 1936, Lawrence authorized the Pelton Company to increase the gap of the magnet another 2.5 inches and to bolt on peripheral iron rings to extend the pole faces. In September 1937 the new 37-inch cyclotron yielded 8 MeV deuterons. The use of rubber gaskets in the 37-inch reduced the need for heroics in leak detection.

Preliminary design for the most costly part of the even larger medical cyclotron, its magnet, was begun on a scale model, rather than, as had been the practice, on the thing itself. From trials with the model, Alvarez concluded that the best magnet for the money would have 60-inch pole pieces. Calculations by Robert R. Wilson removed much of the mystery from the operation of the cyclotron. And most important of all, engineering procedure and standards entered in the person of William Brobeck, a mechanical engineer who volunteered his services on discovering that the Laboratory had never heard of preventive maintenance. Brobeck scaled up the plans from Alvarez's model to the 60-inch machine, the main members of which were built by the Moore Drydock Company in Oakland. Brobeck's intervention introduced the collaboration between engineers and physicists that would henceforth be required in the design of accelerators, just as the commissioning of construction in a shipyard symbolized the initiation of fundamental physics on an industrial scale.

The first external cyclotron beam, obtained on March 26, 1936. The glow arises from the ionization of the air by the 5.8 MeV deuterons.

As the machines became more reliable and demands upon them increased, organization of staff and regulation of experiments became necessary. Up through 1936 the machine ran when it could, tended by a two-man crew in two shifts a day. Every Monday a weekly schedule appeared on which the staff indicated the shifts they preferred and the targets they wanted irradiated; crews were composed primarily of the physicists who had built the machine and ran experiments on it. In 1937 the pace increased: an owl shift (11 in the evening to 3 in the morning) started up in May to make phosphorus-32 for John Lawrence's experiments; in July, primarily to meet demands of biological tracer research, the Laboratory began to work around the clock. Slavery to the machine began to irritate, and the good fellowship that characterized Lawrence's enterprise might have cracked like sealing wax had the war not brought other employment and motivation.

Preoccupation with the machine and isotope manufacture may have cost the Laboratory staff important discoveries in nuclear physics. But they did as well as or better than other accelerator laboratories. Their serfdom kept their apparatus in good order, while physicists who attempted to apply cyclotrons to research as soon and as often as possible found their machines frequently in need of repairs they could not readily make. At Berkeley, however, dependable beams allowed, among much else, McMillan to discover long-lived radionuclides; Alvarez and Felix Bloch to measure the magnetic moment of the neutron; Segre to find technetium; and Alvarez and Robert Cornog to discover the stability of helium-3 and the radioactivity of tritium.

The expertise of Lawrence's "boys," as he liked to call them, drew prospective cyclotron designers to Berkeley from around the world. After a time in Lawrence's school they went forth to multiply machines in the Berkeley style. One of the first envoys was Livingston, who built at Cornell and MIT. Already in 1936 cyclotrons financed by the Research Corporation and the Rockefeller Foundation and designed for physical, chemical, and biological work were under construction at Columbia, Cornell, Princeton, and Purdue, and at the Universities of Chicago, Illinois, Michigan, Pennsylvania, and Rochester. British visitors returned from Berkeley to build in Cambridge, Liverpool, and Birmingham. Other machines came into existence at Harvard, Indiana, and Washington University of St. Louis. In Europe, cyclotrons appeared at Bohr's Institute in Copenhagen, at Joliot's in Paris, and at Siegbahn's in Stockholm. In the French style Joliot had tried to go it alone; but he mismanaged his magnet design and needed help from Berkeley to get his protons spinning. Japanese physicists recreated every detail they had seen in Berkeley, down to stop-gaps and jerry-rigging. Almost all the cyclotron laboratories in existence in 1939 had a direct tie to the first and biggest of them all.

EPISODE 1 A Productive Error