The Age of Radiance (30 page)

At Berkeley, chemist Glenn Seaborg had tried various techniques to extract plutonium from uranium, getting a yield of 250 parts per million, or two tons producing an amount the size of a penny. Testing this required microchemistry, where under 30x power microscopes, capillary straw pipettes became test tubes and a single quartz fiber with a platinum-foil tray was a weighing scale. Next, about twelve miles from K-25, rose the $427 million Y-12 Plant, where thirteen thousand workers ran Lawrence’s Alpha calutrons: 3,000-to-10,000-ton, 250-foot-long magnets across ninety-six vacuum tanks in five merry-go-rounds to separate uranium-235. His next-generation Beta calutrons had twenty-five hundred tanks, eighty-five thousand
vacuum tubes, were the lengths of four football fields, and employed twenty-five thousand.

Building his magnets required five thousand tons of copper, which in wartime couldn’t even be gotten by Leslie Groves, but a substitute was discovered—silver. When Undersecretary of the Treasury Daniel W. Bell was asked for six thousand tons of silver bullion, he erupted, “You may think of silver in tons, but the Treasury will always think of silver in troy ounces!” Eventually, 14,700 tons were needed, costing Groves $300 million. A pickling plant had to be built on-site for cleaning the pipes.

By bending the path of speeding uranium ions, the lighter 235s would pull slightly ahead of the 238s in the proton merry-go-rounds and splat against a target in a slightly different location. This low-production method, with less than 5 percent of the distilled 235 hitting the sweet spot, required constant monitoring of the magnetic pulse to correct for the wavering voltages of the factory’s power supply. At first the calutrons were run by Berkeley scientists to work out the bugs; then they were turned over to local women sitting on stools, most of whom were high school dropouts. The two teams had a race . . . and the women won. The management believed it was because they were
“trained like soldiers not to reason why,” while “the scientists could not refrain from time-consuming investigation of the cause of even minor fluctuations of the dials.”

Calutron operator Theodore Rockwell:
“If you walked along the wooden catwalk over the magnet, you could feel the tug of the magnetic field on the nails in your shoes. It was like walking through glue. People who worked on the calutrons would take their watch into the watchmaker and discover that it was all smashed inside. The magnetic field had grabbed the steel parts and yanked them out by the roots. . . . One time they were bringing a big steel plate in and got too close to the magnetic field. The plate pinned some poor guy like a butterfly against the magnetic field. So the guys ran over to the boss and said, ‘Shut down the magnet! Shut down the magnet! We got to get that guy off.’ And the boss replied, ‘I’ve been told the war is killing three hundred people an hour. If we shut down the magnet, it will take days to get restabilized and get production back up again, and that’s hundreds of lives. I’m not going to do that. You’re going to have to pry him off with two-by-fours.’ Which is what they did. Luckily he wasn’t badly hurt, but that showed what our priorities were.”

By the start of 1944, Urey’s membrane cascades were still corroding viciously. The Naval Research Lab’s Phil Abelson told Groves about their work with thermal diffusion, which had been pioneered by the omnipresent
yet underappreciated fission codiscoverer and U-235 cocalculator, Otto Robert Frisch, and Groves decided to try that method as well. Construction of S-50, a thermal-diffusion plant, started on June 24, 1944, with twenty-one racks of 2,142 forty-eight-foot-tall columns housing three concentric tubes ferrying 545°F steam on the outside, cool water on the inside, and uranium hexafluoride in the middle—the lighter 235 floated up with the rising heat, while the heavier 238 settled down at the colder bottom.

By March 1945, Oppenheimer understood that feeding the various outputs into each other dramatically increased the yield, so the thermal-diffusion output was sent to be gaseously diffused, and those results were in turn racetracked in the calutron merry-go-rounds. The final product was 89 percent enriched—good enough to power the Bomb. Yet, by the summer of 1944, even with this massive effort, the calutrons were producing only enough U-235 to make a single explosive. Because they had so little HEU and the uranium gun bomb was so simple in design, Oppenheimer decided it could be dropped on the enemy without even being tested.

The third Oak Ridge plant, the $12 million pilot nuclear reactor X-10, employed 1,513 people and was built ten miles from Y-12. It was known as the Black Barn and joined Chicago Pile-2 in creating plutonium-239 for the implosion bomb. The Oak Ridge reactor was modified from Fermi’s Chicago pile by using channels for replacing exhausted uranium rods with fresh fuel—the depleted uranium used to make armor-penetrating bullets—and pressurized helium as a coolant. Coolant was a topic sure to cause an argument, with Fermi insisting on air, Szilard wanting liquid bismuth (as he and Einstein had used in their refrigerator patent), and Wigner believing in ordinary water from the river. In time, Szilard’s technique would be used with breeder reactors, and Wigner’s with standard burners.

With Fermi in attendance, X-10 went critical on November 4, 1943. One of its engineers, Arthur Rupp, had not believed in Wigner’s calculations of what would happen, yet the results almost exactly matched the Hungarian’s predictions. “I knew then,” Rupp said, “the atomic bomb was going to work!” X-10 would, for the implosion bomb, produce 326.4 grams of plutonium, and Eugene Wigner would become Oak Ridge’s “patron saint.”

Though the army was fully against it for security reasons, Emilio Segrè insisted on touring Oak Ridge to make sure it was operating properly. He found that the Tennessee reactor employees thought they could use a small amount of water to contain the extract, having no idea the water would then become lethally radiant. They also didn’t realize that storing fuel in adjacent
rooms against a shared wall could start a chain reaction. Segrè was horrified, and Oppenheimer sent in Richard Feynman to do a follow-up inspection. Feynman was told by his boss that, if the army refused to listen to the physicist, he should say, “Los Alamos cannot accept the responsibility for the Oak Ridge plant.” That suitably alarmed the army bureaucrats, and the physicists’ recommendations were followed.

The Oak Ridge reactor was a proving ground for a much bigger operation. The small towns of Hanford, White Bluffs, and Richland in the dry sheep and vineyard inlands of Washington State were evacuated so that fifty-one thousand people—forty-seven thousand of them men—living in a city of barracks and the largest trailer park in history, could build seven nuclear power plants requiring twenty-five thousand gallons a minute of cooling Columbia River water on a half million acres to be known as Hanford Engineer Works. Their bakery made twenty thousand pies a day, and their auditorium seated five thousand for movie nights; the other entertainment was sitting on blankets on the dusty streets, playing cards, gambling, and fighting, the last stopped by security with fire hoses. The work was finished in eighteen months, and their camp was destroyed.

Those going to Hanford imagined it would be like the Washington State of postcards—snow-peaked mountains, crystal mountain streams, great camping, hunting, and fishing. Instead, they arrived at the Columbia Basin desert of sand dunes, saw grass, tumbleweeds, and pygmy rabbits, next to a barren lava plateau—the Scablands. One ordnance émigré from Denver said, “It was so darned bleak. If I’d had the price of a ticket, I wouldn’t have stayed.” The darned bleak was accompanied by dust storms so ferocious they were known as “termination winds” because so many employees gave up and resigned after suffering through one.

On September 13, 1944, Leona Marshall, Crawford Greenewalt, and Enrico Fermi climbed a twelve-story ladder to survey the two-thirds-finished Hanford site, where Fermi would insert the first uranium slug into the first of the three plutonium-generating reactors. It ran perfectly for twelve hours, then the chain faltered and died. The next morning it was back chain-reacting, but twelve hours later it died all over again. Princeton’s John Wheeler theorized that a by-product in the chain was absorbing neutrons, and this turned out to be xenon gas. Wheeler advised DuPont to add more uranium channels, and they tried amending Wigner’s 1,500 with an additional 504. This kept the xenon from overwhelming the uranium, and the plutonium was now generated on schedule.

Hanford operations manager Walter Simon: “Fermi was very discreet
about disagreements. He was a very pleasant person. His mind raced all the time. For instance, if there was a little time to kill while they were loading the reactor, he would do equations in his head, with someone next to him with a calculator. You know, multiply 999 by 62 and divide this by that, and he did that for amusement. His mind raced so much the only way he could relax was to walk on the desert. They would try to take him to a movie, and he would sit there, and in five minutes he would have the whole plot figured out.”

One evening, Sam Allison, Arthur Compton, and Enrico Fermi were taking the train to Hanford, a ride that seemed to go on for a purgatorial eternity. To pass the dead hours, Compton said,
“Enrico, when I was in the Andes Mountains on my cosmic-ray trips, I noticed that at very high altitudes my watch did not keep good time. I thought about this considerably and finally came to an explanation which satisfied me. Let’s hear you discourse on the subject.” Enrico took out his slide rule, a piece of paper, and a pencil and, within a few minutes, had totted up the formula for the interactions of air pressure on a watch balance wheel’s inertia, quickly producing a figure that matched the dissatisfactions of high-elevation timekeeping. Allison said he would never forget Compton’s amazed look.

Shortly after that first Hanford reactor began to be tested early in 1944, a balloon appeared in the sky. It was one of thousands carrying incendiary bombs sent by Japan to incinerate the American West. Though some of these fire balloons did cause forest fires in Northern California, Oregon, and Washington, this one struck the electric line carrying power to the reactor building and shut it down.

By February 4, 1945, Hanford, Oak Ridge, and Chicago reached their target monthly yield of twenty-one kilos of the flourlike yellow-green plutonium oxide, and a few months later, Oak Ridge began sending its HEU to Los Alamos. When Oak Ridge’s U-235 was first unwrapped after reaching the mesa, it was a silvery metal. Contact with the air turned it dawn-sky blue, deepening to cobalt, and then purple. Like the warm, silvery puppy that was plutonium, and the blue-green thrill that was radium, the U-235 also seemed to be, in some way, alive.

Groves now knew that everything was working as he’d hoped and prayed, and that by the end of the year, they would have enough fissile core for eighteen bombs. That is, if the implosion bomb worked at all.

R
obert Serber gave the introductory lecture to new arrivals at Los Alamos and, worried about omnipresent USED construction crews able to eavesdrop
on every meeting, he called the Bomb “the gadget.” Serber estimated critical mass could be arrived at with fifteen kilos (thirty-three pounds) of uranium-235 or five kilos (eleven pounds) of plutonium-239, surrounded by a shell of ordinary uranium. In the two years it would take Hanford, Oak Ridge, and Chicago to amass that quantity of fissile ordnance, Los Alamos would have to design an atomic trigger. Dick Feynman:
“All science stopped during the war except the little bit that was done at Los Alamos. And that was not much science; it was mostly engineering.”

The labors of the mesa were far more dangerous than any civilian knew. The MAUD report included an atom-bomb gun design, which fired a plug of uranium at a bowl of uranium. This would become Little Boy, used against Hiroshima, and was a very basic idea—U-235 molded into six-inch stackable washers with a four-inch hole in the middle that would fit a four-inch-round plug of the same, also made of disks. One design question was, how much of the rare and precious uranium would the Little Boy gun-style bomb need to work? In 1944, Otto Robert Frisch was assigned this investigation, and his method was, by any standards, hair-raising. In Omega Canyon, so remote as to result in as few casualties as possible if the worst happened, Otto built a small tower, the “guillotine,” which dropped plugs of uranium metal through blocks of uranium metal surrounded by tamper—the basic design for the Hiroshima bomb—which went supercritical for less than a second and immediately showed the quantity needed. Dick Feynman said it “was like tickling the tail of a sleeping dragon . . . as near as we could possibly go towards starting an atomic explosion without actually being blown up.” After testing with Frisch’s guillotine, the amount of each segment could be shimmed by adding additional washers of isotope. In the bomb, a gunlike explosion of cordite would unite the two and could be triggered by pressure or proximity fuse or just about anything else in the Pentagon trigger basket. Little Boy was easy to set off accidentally, such as by dropping it on its nose, so small studs were put in place that would be sheared away by the cordite blow. The design was so foolproof that Oppenheimer didn’t bother testing it before it was dropped over Japanese skies, and it was considered so dangerously easy for a foe to build that, after Japan’s surrender, the design files were destroyed in a Los Alamos bonfire.