* * *
Twelve days after Enrico Fermi proved the chain reaction in Chicago on December 2, 1942, Groves had assembled a list of criteria for a plutonium production area and definitely and finally ruled out Tennessee. “The Clinton site . . . was not far from Knoxville,” he comments, “and while I felt that the possibility of serious danger was small, we could not be absolutely sure; no one knew what might happen, if anything, when a chain reaction was attempted in a large reactor. If because of some unknown and unanticipated factor a reactor were to explode and throw great quantities of highly radioactive materials into the atmosphere when the wind was blowing toward Knoxville, the loss of life and the damage to health in the area might be catastrophic.” Such an accident might “wipe out all semblance of security in the project,” Groves could imagine, and it might render the electromagnetic and gaseous-diffusion plants “inoperable.”1893 Better to site plutonium production somewhere far away.
The production piles needed plentiful electricity and water for blowing and cooling the helium that was planned to cool them. For safety they needed space. Those criteria suggested the great river systems of the Far West, particularly the Columbia River basin. Groves sent out an officer who would administer the plutonium reservation along with the civilian engineer who would supervise construction for Du Pont. Besides picking the site he wanted the two men to get used to working together. They did, agreeing on a promising location in south-central Washington State, and arrived back in Groves’ office on New Year’s Eve to report. The general received a real estate appraisal on January 21, 1943.1894 By then he had already personally walked the ground.
Eastward of the Cascade Range, twenty air miles east of the city of Yakima, the blue, cold, fast-running Columbia River bends east, then northeast, abruptly ninety degrees southeast and finally due south through a flat, arid scrubland on its last excursion toward the continental interior before it makes its great bend below Pasco to course directly westward two hundred fifty miles to the sea. Even that far inland the river is wide and deep and veined in season with salmon, but the sandy plain surrounding wins little of the river’s water and the barrier of the Cascades denies it more than six inches a year of rain.
The site Groves’ representatives discovered, and Groves acquired at the end of January at a cost of about $5.1 million, was contained within the eastward excursion of the Columbia: some 500,000 acres, about 780 square miles, devoted primarily to sheep grazing but varied with a few irrigated orchards and vineyards and a farm or two thriving in wartime on irrigated crops of peppermint. Temperatures ranged from a maximum of 114° in the long, dry summers to rare −27° winter lows. Roads were sparse on the roughly circular thirty-mile tract. A Union Pacific railroad line crossed one corner; a double electric power line of 230 kilovolts traversed the northwest sector on its way from Grand Coulee Dam to Bonneville Dam. Gable Mountain, an isolated basalt outcropping that rose five hundred feet above the sedimentary plain a few miles southwest of the ninety-degree river bend, divided the riverside land at the bend from the interior. Midway down the tract where a ferry crossed the Columbia, a half-abandoned riverside village, population about 100, supplied a base of buildings and gave the Hanford Engineer Works its name.1895
Groves could hardly build Hanford until he knew more about the plant that would go there. It was clear that he would need enormous quantities of concrete to shield the production piles and chemical processing buildings; his Hanford engineer searched out accessible beds of gravel and aggregate to quarry. An accident might release radioactivity into the air; that called for thorough meteorological work. The river water needed study; so did the river’s valuable salmon, to see how they would take to mild doses of transient radioactivity from pile discharge flow. Roads had to be paved, power sources tapped, hutments and barracks built for tens of thousands of construction workers.
What had come up once again for discussion early in 1943 was how the plutonium production piles—the Du Pont engineers were beginning to call them reactors—should be cooled. Crawford Greenewalt, in charge of plutonium production for Du Pont, continued to plan for helium cooling because the noble gas had no absorption cross section at all for neutrons. But it would need to be pumped through the piles under high pressure; that would require large, powerful compressors Greenewalt was not at all sure he had time to build. Enormous steel tanks would be needed to contain the gas; they would have to allow access to the pile but still remain airtight, a formidable challenge to engineer or even simply to weld.
Eugene Wigner came to the project’s rescue. Fermi had found k for CP-1 higher than he expected. The Stagg Field pile had been assembled largely from uranium oxide. Its graphite had varied in quality, improving along the way. A production pile of pure uranium metal and high-quality graphite would find k higher yet—high enough, Wigner calculated, to make water cooling practical.
Wigner’s team designed a 28- by 36-foot graphite cylinder lying on its side and penetrated through its entire length horizontally by more than a thousand aluminum tubes. Two hundred tons of uranium slugs the size of rolls of quarters would fill these tubes. Chain-reacting within 1,200 tons of graphite, the uranium would generate 250,000 kilowatts of heat; cooling water pumped through the aluminum tubes around the uranium slugs at the rate of 75,000 gallons per minute would dissipate that heat. The slugs would not go naked into the torrent; Wigner intended that they also should be separately sheathed in aluminum—canned. When they had burned long enough—100 days—to transmute about 1 atom in every 4,000 into plutonium the irradiated slugs could be pushed out the back of the pile simply by loading fresh slugs in at the front.1896, 1897 The hot slugs would fall into a deep pool of pure water that would safely confine their intense but short-lived fission-product radioactivity. After 60 days they could be fished out and carted off for chemical separation.
The Wigner design was elegantly simple. Greenewalt saw engineering problems—in particular the question whether corrosion of the aluminum tubes would block the flow of cooling water—and studied helium and water side by side until the middle of February. Corrosion studies were promising. “With water of high purity,” writes Arthur Compton, “the evidence indicated that no serious difficulties from this source should arise.”1898 Greenewalt opted then for water cooling. Wigner, whom Leo Szilard calls “the conscience of the Project from its early beginnings to its very end,” who worried constantly about German progress, wondered angrily why it had taken Du Pont three months to see the value of a system he and his group had judged superior in the summer of 1942.1899
With that basic decision construction could begin at Hanford. Three production piles would go up at six-mile intervals along the Columbia River, two upstream and one downstream of its ninety-degree bend. Ten miles south, screened behind Gable Mountain, Du Pont would build four chemical-separation plants paired at two sites. The former town of Hanford would become a central construction camp serving all five construction areas.
The work proceeded slowly, dogged by recruiting problems. The nation at war had moved beyond full employment to severe labor shortages and men and women willing to camp out on godforsaken scrubland far from any major city were hard to find. Frequent sandstorms plagued the area, writes Leona Woods, now Leona Marshall after marrying fellow physicist John Marshall of Fermi’s staff. “Local storms were caused by tearing up the desert floor for roads, and construction sites were suffocating. Wind-blown sand covered faces, hair, and hands and got into eyes and teeth. . . . After each storm, the number of people quitting might be as much as twice the average. When the storms were at their worst, buses and other traffic came to a stop until the roads were visible through the greyblack clouds of dust.”1900 Stoics who stayed on called the dust “termination powder.”
“The most essential thing to bring with you is a padlock,” a project recruiting pamphlet ominously announced. “The next important things are towels, coat hangers and a thermos bottle. Don’t bring cameras or guns.”1901 Hanford, says Marshall, “wa
s a tough town. There was nothing to do after work except fight, with the result that occasionally bodies were found in garbage cans the next morning.”1902 Du Pont built saloons with windows hinged for easy tear-gas lobbing. Eventually some 5,000 construction workers struggled in the desert dust and Du Pont built more than two hundred barracks to house them. Meat rationing stopped at the edge of the reservation; there were no meatless Tuesdays in the vast Hanford mess halls, a significant enticement for recruiting. The gray coyotes of the region fed sleek in turn on rabbits killed by cars and trucks driving the new reservation roads.
By August 1943 work had begun on the water-treatment plants for the three piles, capacity sufficient to supply a city of one million people. Du Pont released pile-design drawings in Wilmington, Delaware, on October 4 and the company’s engineers staked out the first pile, 100-B, beside the Columbia on October 10. After excavating, reports an official history, “work gangs began to lay the first of 390 tons of structural steel, 17,400 cubic yards of concrete, 50,000 concrete blocks, and 71,000 concrete bricks that went into the pile buildings.1903 Starting with the foundations for the pile and the deep water basins behind it where the irradiated slugs would be collected after discharge, the work crews were well above ground by the end of the year.” The forty-foot windowless concrete monolith they were building was hollow, however: installation of B pile would not begin until February 1944.1904
* * *
“There was a large change of scale from the Chicago to the Hanford piles,” Laura Fermi remarks. “As Fermi would have put it, they were different animals.”1905 So also were Ernest Lawrence’s behemoth mass spectrometers and John Dunning’s gaseous-diffusion factory with its 5 million barrier tubes. The mighty scale of the works at Clinton and Hanford is a measure of the desperation of the United States to protect itself from the most serious potential threat to its sovereignty it had yet confronted—even though that threat, of a German atomic bomb, proved to be an image in a darkened mirror. It is also a measure of the sheer recalcitrance of heavy-metal isotopes. Niels Bohr had insisted in 1939 that U235 could be separated from U238 only by turning the country into a gigantic factory. “Years later,” writes Edward Teller, “when Bohr came to Los Alamos, I was prepared to say, ‘You see . . .’ But before I could open my mouth, he said, ‘You see, I told you it couldn’t be done without turning the whole country into a factory. You have done just that.’ ”1906
The monumental scale reveals another desperation as well: how ambitiously the nation was moving to claim the prize. And to deny it to others, even to the British until Winston Churchill turned Franklin Roosevelt’s head at the conference in Quebec in August 1943, where Operation Overlord, the 1944 invasion of Europe across the beaches of Normandy, was planned. Before then, in June, Groves had demonstrated this last desperation at its most overweening: he proposed to the Military Policy Committee that the United States attempt to acquire total control of all the world’s known supplies of uranium ore. When the Union Minière refused to reopen its flooded Shinkolobwe Mine in the Belgian Congo, Groves had to turn to the British, who owned a significant minority interest in the Belgian firm, for help; after Quebec the partnership evolved into an agreement between the two nations known as the Combined Development Trust to search out world supplies. That uranium is common in the crust of the earth to the extent of millions of tons Groves may not have known. In 1943, when the element in useful concentrations was thought to be rare, the general, acting on behalf of the nation to which he gave unquestioning devotion, exercised himself to hoard for his country’s exclusive use every last pound. He might as well have tried to hoard the sea.
* * *
Work toward an atomic bomb had begun in the USSR in 1939. A thirtysix-year-old nuclear physicist, Igor Kurchatov, the head of a major laboratory since his late twenties, alerted his government then to the possible military significance of nuclear fission. Kurchatov suspected that fission research might be under way already in Nazi Germany. Soviet physicists realized in 1940 that the United States must also be pursuing a program when the names of prominent physicists, chemists, metallurgists and mathematicians disappeared from international journals: secrecy itself gave the secret away.1907
The German invasion of the USSR in June 1941 temporarily ended what had hardly been begun. “The advance of the enemy turned everyone’s thoughts and energies to one single job,” writes Academician Igor Golovin, a colleague of Kurchatov and his biographer: “to halt the invasion. Laboratories were deserted. Equipment, instruments and books were packed up, and valuable records shipped east for safety.”1908 The invasion rearranged research priorities. Radar now took first place, naval mine detection second, atomic bombs a poor third. Kurchatov moved to Kazan, four hundred miles east of Moscow beyond Gorky, to study defenses against naval mines.
In Kazan at the end of 1941 he heard from George Flerov, one of the two young physicists in his Moscow laboratory who had discovered the spontaneous fission of uranium in 1940 and reported their discovery in a cable to the Physical Review.1 Flerov had attended an international meeting of scientists in Moscow in October and heard Peter Kapitza, Ernest Rutherford’s protege, when asked what scientists could do to help the war effort, respond in part:
In recent years a new possibility—nuclear energy—has been discovered. Theoretical calculations show that, if a contemporary bomb can for example destroy a whole city block, an atomic bomb, even of small dimensions, if it can be realized, can easily annihilate a great capital city having a few million inhabitants.1909
Thus recalled to their earlier work, Flerov challenged Kurchatov as he had already in a similar letter challenged the State Defense Committee that “no time must be lost in making a uranium bomb.”1910 The first requirement was fast-neutron research, he wrote. The MAUD Report had only just made that necessity clear to the United States.
Kurchatov disagreed. Research toward a uranium weapon seemed too far removed from the immediate necessities of war. But the Soviet government in the meantime had assembled an advisory committee that included Kapitza and the senior Academician Abram Joffe, Kurchatov’s mentor. The committee endorsed atomic bomb research and recommended Kurchatov to head it. Somewhat reluctantly he accepted.
“So it was that from early 1943 on,” writes his colleague A. P. Alexandrov, “work on this difficult problem was resumed in Moscow under the leadership of Igor Kurchatov.1911 Nuclear scientists were recalled from the front, from industry, from the research institutes which had been evacuated to the rear. Auxiliary work began in many places.” Auxiliary work included building a cyclotron. Kurchatov moved his institute out of the Soviet capital to an abandoned farm near the Moscow River in the summer of 1943. An artillery range nearby offered an area for explosives testing; “Laboratory No. 2” would be the Soviet Union’s Los Alamos. By January 1944 Kurchatov had assembled a staff of only about twenty scientists and thirty support personnel. “Even so,” writes Herbert York, “they did experiments and made theoretical calculations concerning the reactions involved in both nuclear weapons and nuclear reactors, they began work designed to lead to the production of suitably pure uranium and graphite, and they studied various possible means for the separation of uranium isotopes.”1912 But the Soviet Bear was not yet fully aroused.
* * *
“The kind of man that any employer would have fired as a troublemaker.” Thus Leslie Groves described Leo Szilard in an off-the-record postwar interview, as if the general had arrived first at fission development and Szilard had only been a hireling.1913 Groves seems to have attributed Szilard’s brashness to the fact that he was a Jew. Upon Groves’ appointment to the Manhattan Project he almost immediately judged Szilard a menace. They proceeded to fight out their profound disagreements hand to hand.
The heart of the matter was compartmentalization. Alice Kimball Smith, the historian of the atomic scientists whose husband Cyril was associate division leader in charge of metallurgy at Los Alamos, defines the background of the conflict:
&n
bsp; If the Project could have been run on ideas alone, says Wigner, no one but Szilard would have been needed. Szilard’s more staid scientific colleagues sometimes had trouble adjusting to his mercurial passage from one solution to another; his army associates were horrified, and to make matters worse, Szilard freely indulged in what he once identified as his favorite hobby—baiting brass hats. General Groves, in particular, had been outraged by Szilard’s unabashed view that army compartmentalization rules, which forbade discussion of lines of research that did not immediately impinge on each other, should be ignored in the interests of completing the bomb.1914
The issue for Szilard was openness within the project to facilitate its work. “There is no way of telling beforehand,” he wrote in a 1944 discussion of the problem, “what man is likely to discover and invent a new method which will make the old methods obsolete.”1915 The issue for Groves, to the contrary, was security.
At first Szilard bent the rules and Groves threatened him. In late October 1942, while Fermi moved toward building CP-1, Szilard apparently badgered the Du Pont engineers who arrived in Chicago to take over pile design. Arthur Compton saw this activity as obstructive but not necessarily subversive; on October 26 he wired Groves that he had given Szilard two days TO REMOVE BASE OF OPERATIONS TO NEW YORK. ACTION BASED ON EFFICIENT OPERATION OF ORGANIZATION NOT ON RELIABILITY. ANTICIPATE PROBABLE RESIGNATION.1916 Compton did not know his man. Szilard would not resign, for the simple reason that he believed he was needed to help beat Germany to the bomb. Compton proposed surveillance: SUGGEST ARMY FOLLOW HIS MOTIONS BUT NO DRASTIC ACTION NOW. Two days later Compton hurriedly wired Groves to desist: SZILARD SITUATION STABILIZED WITH HIM REMAINING CHICAGO OUT OF CONTACT WITH ENGINEERS. SUGGEST YOU NOT ACT WITHOUT FURTHER CONSULTATION CONANT AND MYSELF.1917