Now the entire governmental machine began to get to work on the effort, code-named the Manhattan Project, after the headquarters of the Army Corps of Engineers’ district tasked to manage it. Bush appointed Conant to oversee the scientific project from Washington and gave Compton responsibility for academic research throughout the country. Bush also made clear the government’s intent to maintain authority over the project and to transfer it to the army’s control when large-scale production of fissionable materials became necessary. His reasons were simple: Bush knew the money was running out from sources at his disposal and much more was going to be needed. By bringing in the military, he could conceal the project’s costs within the Army Corps of Engineers’ enormous appropriation under line items dubbed “Procurement of New Materials” and “Expediting Production.” Roosevelt did not want to have to justify the Manhattan Project on the Hill. This might slow down the project and jeopardize its secrecy. 31
Many of the physicists who would soon be brought into the Manhattan Project were refugees, recent immigrants to the United States. This was partly because they included some of the world’s best physicists, but there was another reason as well: many native-born American physicists had been swept up earlier in military research on radar and the proximity fuse, which appeared to have a more immediate military application to Allied success in the war. As a result, refugees were the main remaining source of available scientific brainpower to work on the project. The very restrictions and limitations imposed upon refugee scientists—which had delayed the government’s embrace of the project—facilitated their leading roles in the bomb’s development once the government decided to support it. 32 This irony would have a significant, if unstated, impact down the road, when disputes arose about the long-term political consequences of what the scientists and the government were doing.
In the end, the refugee physicists and their native-born colleagues did not protest their loss of control over the project in December 1941. Most of them, in fact, welcomed it because they thought it would insulate them from political pressure and criticism. Their acceptance of this condition was the tacit price of their admission into the project. It was also a measure of their loyalty by those at the top. “I think [Ernest Lawrence] now understands this,” Bush said, “and I am sure Arthur Compton does, and I think our difficulties in this regard are over.” 33 The government was giving physicists, whom Bush and others in top councils considered “somewhat naive and lacking in discretion,” 34 the responsibility for making an atomic bomb, not for helping to decide how it would be used.
Oak Ridge was a remote rural area surrounding the Clinch River eighteen miles from Knoxville, Tennessee. It was beautiful country, rolling hills dotted with dogwood, oak, and pine trees, and situated between the Great Smoky Mountains to the east and the Cumberland Mountains to the west. It answered all the requirements for a sprawling plant to separate U-235 isotopes: an isolated area in the midst of the vast power grid of the Tennessee Valley Authority, an abundant water supply, relatively few people to relocate, good access by road and train, and a mild climate that permitted outdoor work the year round. Here on a 59,000-acre site, 32,000 construction workers built and 47,000 operating personnel maintained a gigantic forty-two-acre separation plant flanked by facilities covering some fifty additional acres and containing more than six thousand miles of pipe that was the largest factory complex on earth when it was finished.
U-235 was separated at Oak Ridge by three different methods—no one knew which would prove most effective. The first method was electromagnetic separation, using giant cyclotrons designed by Ernest Lawrence. Uranium atoms were stripped of electrons in a vacuum. Then they were electrically charged and thus made more susceptible to outside magnetism. The heavier U-238 was more sluggish, so the lighter U-235 could gradually and painstakingly be separated out. The enormous separation chambers contained vacuum pumps, more powerful than any ever built, that pushed through millions of gallons of oil a day; the magnet coil windings required 27,750,000 pounds of silver (the metal, worth $400 million at 1940s prices, was borrowed from the Treasury Department).
The second method was gaseous diffusion, developed by Columbia University physicists Harold Urey and John Dunning. When ordinary uranium was mixed with fluorine, the resulting compound—uranium hexafluoride—was a gas. When the uranium hexafluoride gas was forced through the microscopic membrane holes of a filter (or “barrier,” as it was also called), the lighter U-235 passed through faster and the gas on the far side was marginally enriched with the desired isotope. When the process was repeated, the proportion of U-235 increased a little more. Bomb-grade uranium—containing 90 percent U-235—required thousands of passes through the filters.
The third method of U-235 separation was thermal diffusion, pioneered by a former student of Lawrence’s working at the Naval Research Laboratory named Philip Abelson. The apparatus was simple. Long, vertical, concentric pipes were enclosed in cylinders that resembled a gigantic church organ. Each cylinder was composed of a thin nickel pipe within a copper pipe. These two pipes, in turn, were encased in a third one made of galvanized iron. When uranium hexafluoride gas was passed between the hot nickel pipe and the cool copper pipe, the lighter U-235 concentrated near the hot nickel wall and moved upward, while the heavier U-238 moved downward along the cool copper wall. The enriched uranium was then skimmed off at the top. Thermal diffusion could increase the percentage of U-235 in natural uranium by only a small amount, but the enrichment was sufficient to supplement the gaseous diffusion method as another source of material for the electromagnetic racetracks, whose efficiency soared tremendously when fed with even slightly enriched uranium.
The names of the processing plants at Oak Ridge sounded like the combination to a safe: X-10, Y-12, S-50, K-25. All plants except X-10, a plutonium research lab, performed the same function: extracting precious U-235 from U-238. At S-50, thermal diffusion was employed; at Y-12, electromagnetic separation was applied; at K-25, the process was gaseous diffusion. K-25 was the largest building ever constructed up to that time. It was a sight to behold. Spread over 2 million square feet, the U-shaped structure was half a mile long and four hundred feet wide on each side. It was so vast that foremen rode bikes from one part of the building to another. Twelve thousand people, working in three shifts, kept K-25 running day and night, seven days a week. When it was operating, a continuous hum—a high-pitched sound resembling the buzzing of a bee swarm—came from the plant, mixing weirdly with the noises from the nearby woods. The electricity for these mammoth facilities came from the nearby TVA and an on-site powerhouse that was the largest power installation ever built. By war’s end, Oak Ridge would be consuming the equivalent of the total power output on the American side of Niagara Falls—or one-seventh of all electricity generated in the United States.
Lawrence toured the sprawling complex as it was being built, and thrilled at the spectacle. “What you’re doing here is very important,” he told construction workers assembled to hear him give a pep talk. Oak Ridge was a realization of his vision of big physics, and it made him feel proud—like King Henry V addressing his troops before the Battle of Agincourt. “A hundred years from now, people may not remember that there was a war on now,” he told them, emotion rising in his voice, “but they will remember what you were doing.” 35 Privately, Lawrence was awed by what lay ahead. “When you see the magnitude of that operation there,” he wrote after returning to Berkeley, “it sobers you up and makes you realize that whether we want to or not, we’ve got to make things go. We must do it!” 36
The magnets of the cyclotrons that Lawrence had built at Oak Ridge were 250 feet long, and each contained thousands of tons of steel. They were a hundred times larger than the magnet of the 184-inch Berkeley cyclotron—previously the largest in the world. Their magnetic field was so strong that a wrench would be wrested from a workman’s hand, or if he held onto it, he would be pulled against the magnet. But the U-235 separated by these giant cyclotrons o
ffered itself up in only minuscule quantities. Yields were so low from the tons of uranium ore being processed that workers carefully plucked mere specks from their white overalls with tweezers. There were times when they got down on their hands and knees to look for tiny bits of the precious fissionable material.
In south-central Washington State, the small town of Hanford sat in the midst of a vast area of sagebrush and sand, twenty miles north of Richland, bounded on three sides by a huge bend in the Columbia River. This unusual combination—large amounts of water flowing through sparsely peopled desert—made the site suitable for another prong of the Manhattan Project. The Columbia River would provide the enormous amount of water necessary to cool three gigantic piles to be built there for the production of plutonium, an alternative (and more easily obtained) source of fissionable material for the bomb. The isolated location—the population density was just 2.2 persons per square mile—would mitigate the effects of any accidental radioactive release and be easy to guard. In time, the Hanford facility would grow to more than 428,000 acres—500 square miles. 37
To recruit a massive labor force of construction workers for the Hanford site at a time when every war industry in America was begging for manpower was an extraordinary task. The White House cabled regional employment offices, giving preference by direction of the president himself to the Hanford Engineering Works, as it was called, and authorizing them, if necessary, to draft workers from the aircraft industry. In some towns of the Northwest, clergymen were asked to promote Hanford from the pulpit. Veterans of many big public works projects—men who had helped construct huge dams and power plants—had never seen so many people working in the same place at the same time. Living in barracks and trailer camps, they created a massive, sprawling physical plant. The statistics were stunning: 540 structures, more than 600 miles of blacktop, 158 miles of track. Eventually, 132,000 workers (working 126 million man-hours) signed on—nearly as many as had labored to build the Panama Canal. Hanford soon became the fourth-largest city in the state of Washington. The ultimate price of Hanford would reach $358 million, or nearly $5 billion in 2004 dollars. 38
Hanford’s three all-important piles—each one processed two hundred tons of uranium for two hundred days—produced the plutonium, but equally important were the chemical-separation plants that treated the uranium slugs irradiated in the piles. These slugs were so radioactive when they came out that they glowed. 39 Three chemical-separation plants were built in isolated and heavily guarded desert areas south of nearby Gable Mountain. For safety reasons—the plutonium in the irradiated slugs was also highly radioactive—the plants were placed ten miles from the nearest pile and well apart from one another. No one wanted to discover an atomic blast by accident.
The separation plants were sinister-looking, windowless structures with walls eight feet thick—in effect, huge concrete coffins eight hundred feet long and eighty feet wide. Each contained an underground row of forty cells where the irradiated slugs were processed. The operating gallery that ran above the cell rows was a silent, deadly radioactive tunnel with glaring electric lamps, where no human being could survive. Because of plutonium’s deadly toxicity, metallurgists had to be specially trained to handle it. They wore rubber gloves, worked behind protective shielding, and manipulated the plutonium with long tongs. Not only was the air filtered and ventilated, but a microscopist was hired to analyze its dust. In these gargantuan coffin- like structures, workers operating remote-control machinery around the clock tortuously squeezed out plutonium in a concentration of about 250 parts per million, a half-pound radioactive pellet from every ton of irradiated uranium.
To build a bomb from materials that didn’t yet exist in measurable quantities, involving the commitment of an extraordinary range of human and material resources, in the midst of a global war—it was an improbable undertaking. Yet out of nothing would be created a vast industrial enterprise. Bohr’s prediction had not been far off the mark: before the war was over, the Manhattan Project would consume more than $2 billion, employ 500,000 people directly and indirectly, and mobilize vast material resources. The project exemplified human ingenuity and determination, an immense undertaking into which industrial power was harnessed at vast cost and extraordinary effort. There was something vitally American about the Manhattan Project: no other nation in a world at war had the time and money to attempt such a thing.
An all-out race to build an atomic bomb was now under way after a delay of more than two years. Scientists entered the race—against German scientists believed to have a two-year head start—convinced that the outcome of the war depended on their ability to recover lost time. For them, the bomb’s rapid development was the single most important necessity of the war. It was a matter of survival.
CHAPTER 4
The Met Lab
SUNDAY, DECEMBER 7, 1941, found Arthur Compton on the morning train from Washington to New York. At the Wilmington, Delaware, station a passenger boarded his compartment and shouted nervously that the Japanese had attacked the U.S. Pacific Fleet at Pearl Harbor, bringing America into the war at last. The news sent Compton into a reverie as the train left Wilmington and raced north across the lush pastureland of rural New Jersey. Compton could see the neo-Gothic towers of Princeton University off in the distance. There was the Graduate College, where he had lived, and the Palmer Physical Laboratory, where he had conducted experiments as a graduate student two decades before. How different those buildings now seemed to Compton. They were still outwardly peaceful ivory towers, but Compton knew that within their walls were active and creative minds working on an immensely destructive weapon that might decide the outcome of the war that America had just entered. He felt conflicted about his own work on an immensely destructive weapon because of a pacifist upbringing by his Mennonite mother. And yet he felt he had no other choice, especially now.
When Compton reached New York that afternoon, he took a taxi from Penn Station up to Columbia University, where he met with Rabi, Szilard, and Fermi to discuss producing plutonium in a chain-reacting pile. This was crucial because producing plutonium by bombarding uranium in a cyclotron—what had been done at the Rad Lab—was not a practical method: at Berkeley, a kilogram of uranium bombarded for a week produced less than a millionth of a gram of plutonium. At that rate, a yearlong bombardment might produce fifty micrograms (smaller than a single grain of sand); it would take 20,000 years to make a kilogram—much better than Bohr’s early estimate of 26,445 years to produce one gram, but still nowhere close to what was needed. No one knew exactly how much plutonium would be necessary for a bomb, but estimates ran to several kilograms. Szilard and Fermi expressed confidence that a chain-reacting uranium-graphite pile would be a feasible method of producing large amounts of plutonium.
Until the War Department took control of the Manhattan Project in the fall of 1942, Compton was the de facto leader. Project research was then under way at universities scattered across the country. Compton thought it all should be centralized in one location to avoid duplication, ease communication, and save precious time. He called a meeting in Chicago in late January 1942 to decide where. Compton had the flu, and ran the meeting from a sickbed in one of the spare bedrooms on the third floor of his house. Szilard and Lawrence were both there. “Each was arguing the merits of his own location,” Compton later wrote, “and every case was good. I presented the case for Chicago.”
First, Compton stressed that he already had the support of the University of Chicago. “We will turn the university inside out if necessary to help win this war,” its vice-president had told him. 1 Second, more scientists were available in the Midwest than on the coasts, where universities had already been drained for other war work. Finally, Chicago was conveniently and centrally located for travel to other sites.
“You’ll never get a chain reaction going here,” scoffed Lawrence. “The whole tempo of the University of Chicago is too slow.” He argued in favor of moving everyone to Berkeley. Compton had known Lawrence s
ince Lawrence was a graduate student at Chicago and Compton was chairman of its physics department and a dean; he had no intention of now becoming Lawrence’s subordinate. “We’ll have the chain reaction going here by the end of the year,” Compton bristled. Needing to make a decision, he announced that Chicago would be the site. The decision was logical. University administrators had promised unlimited support, which would be necessary—whole buildings would have to be turned over to researchers. Chicago’s location in the center of the country also made it a compromise relocation site for scientists on both coasts. 2
Lawrence told Compton that plutonium would be highly radioactive and thus dangerous, but the chemical-separation work could be done. He promised to commit the Rad Lab’s best chemists to the job, perfecting the process in the lab and then applying it on an industrial scale at Hanford. Armed with this information, Compton laid out an ambitious timetable for Washington: “By July 1, 1942, to determine whether a chain reaction was possible. By January 1943, to achieve the first chain reaction. By January 1944, to extract [plutonium] from uranium. By January 1945, to have a bomb.” 3
Szilard expressed his commitment to the project, rather than to a specific location. However, he opposed moving to the Midwest because he would be isolated from physicists on both coasts. Better, he thought, to center the effort in the East—or shift it to Berkeley. Additionally, Szilard said that Compton should pay particular heed to Fermi, who would have to “overcome his strong preference for Columbia.” “It would be obviously wrong,” he said, “to decide in favor of a place as long as Fermi had a strong objection.” 4
Pandora's Keepers Page 9
