The Girls of Atomic City (16 page)

Between there and Los Alamos, Tubealloy might be chloridized, oxidized, sublimized, fluorized, vaporized, bombarded, spun, separated, weighed, assessed, assayed, measured, and measured again, its very characteristic scrutinized at each stage of processing.

The challenge was not securing enough
raw
Tubealloy; it was transforming that raw Tubealloy into fuel for two different models of the Gadget. One version of the Gadget would use
enriched
Tubealloy: Tubealloy with a high concentration of isotope 235. The second version of the Gadget used 49 for fuel, an extremely powerful and toxic by-product of Tubealloy fission.

Three principal Project sites were engaged in around-the-clock work to tackle this task of creating and fueling the Gadget, with the help of other facilities and companies across the country. Gadget design and assembly took place at Site Y in Los Alamos, New Mexico, on the site of a former boys’ school. Principal production of 49 went on at Site W in Washington State. Site X, Clinton Engineer Works in Tennessee, would eventually house four different plants dedicated to handling Tubealloy.

SCALING UP

In February 1943, construction of the plants code-named Y-12 and X-10 began at the Clinton Engineer Works. X-10 was a pilot reactor, but much bigger than the pile created by Fermi’s team in Chicago. Using slugs of Tubealloy, canned by the Aluminum Company of America, X-10 produced 49 by means of a fission chain reaction. As Tubealloy gives up its neutrons, not only is energy released, but those free neutrons split neighboring atoms, which in turn release their neutrons, and so on. But some of those neutrons are
captured
by other atoms. When this happens, the eventual product is 49.

On November 4, 1943, X-10 went “critical”: The chain reaction of neutrons splitting other atoms was self-sustaining. Enrico Fermi and Arthur Compton had traveled to CEW—under the names Henry Farmer and Arthur Holly—for the anticipated event and were yanked from their beds at the Guest House in time for the momentous occasion. The larger scale reactors at Site W in Washington were based on the success of X-10. The remaining three plants at CEW—Y-12, K-25, and S-50—were dedicated to
enriching
Tubealloy alone: separating the Tubealloy 235 from the Tubealloy 238.

The success of the Project hung on the three little neutrons that differentiated T-235 from T-238. Tubealloy 238 is more commonly found in nature. But it’s not as fissionable as the less frequently found Tubealloy 235. In fact, only about seven out of every 1,000
atoms
of Tubealloy are isotope 235. So for every 1,000 pounds of Tubealloy (T), only
seven
are T-235. Imagine having 1,000 grains of rice and only seven of those would cook the way you want.

The Gadget would only get cooking with those precious little grains of T-235. That was the sole purpose of these gargantuan plants: ferreting out those rare and valuable atoms that were needed to fuel the Gadget.

THREE PLANTS: THREE DIFFERENT METHODS

Separating T-235 from T-238 had to be done
physically,
exploiting the minuscule difference in mass between the two. Each plant at CEW used a different approach to—hopefully—achieve this: Y-12 via electromagnetic separation; K-25 via gaseous diffusion; and S-50 via liquid thermal diffusion.

By the summer of 1944, the Project’s expenses were roughly $100 million a month and construction was about halfway done at the mammoth K-25 plant. Situated on a 5,000-acre site, site preparation for the K-25 plant had begun in June 1943, construction in late September. Construction continued at a breakneck pace at the half-mile-long structure, but the General still didn’t know when he would be able to flip the “on” switch.

K-25

The plant’s soaring cooling towers, rising like Appalachian skyscrapers above the secret site, recirculated a volume of water that could serve a city of five million. When finished, the plant would have four stories, more than 44 acres of floor space under a single roof (more than 44 football fields). This made it the largest building of its kind in the world—though precious few, including those who lived in the vicinity, knew it was there.

K-25 used gaseous diffusion to separate T-235 from T-238. Such a method had never been tried on this scale. Here’s how it works:

Tubealloy is converted into a gas, TFL6, which is then pumped through a series of tubes comprised of what the Project called a “barrier.” The barrier was a thin, porous sheet of metal, with identically sized, submicroscopic openings. These openings were incredibly small—numbering in the hundreds of millions per square
centimeter.
The barriers were rolled into tubes, which were sealed in another, larger airtight pipe. The idea was that as a high-pressure stream of Tubealloy gas was pumped through the barrier tubes, more of the lighter 235 would diffuse through the barrier while more of the chunkier 238 would not. As the Tubealloy traveled through a series of these tubes, the lighter 235 would head up the cascade of tubing to the next highest level.

One pass through this process would not be enough—closer to 3,000 stages were needed, accounting for K-25’s gigantic U-shape. (It was as though the shape of the building itself announced to anyone flying overhead what was being processed inside.) Tubealloy would pump through stage after stage, snaking its way around the giant U on a mile-long trip, becoming more highly enriched as it did. That was the idea, anyway.

There was a problem: Design of the barrier still eluded the Project’s scientists. The first barrier tested in a lab had been about the size of a silver dollar, but K-25 would need
acres
of barrier material. So while scientists labored frantically for a solution, workers kept building K-25. Pipes and other structural elements arrived, were inspected and put in place where possible. A company called Midwest Piping had produced three million feet of nickel-plated pipe, and all of it had to be both leakproof and able to resist the corrosive effects of the Tubealloy gas. Standard welding would not suffice—new techniques were developed; a school was set up to train employees. Every detail was crucial, right down to every last spot weld on every single pipe.

Y-12

Meanwhile, in another area of CEW, the Y-12 plant remained the only fully operational option for enriching Tubealloy, ground zero for the electromagnetic separation process that Ernest Lawrence had developed at Berkeley’s Rad Lab. This process centered on calutrons—
U
niversity of
Cal
ifornia Cyclo
trons.
If Tubealloy was the lifeblood of the Project, then Y-12’s calutrons were CEW’s beating heart and soul.

There were Alpha and Beta calutrons, which differed primarily in their size and the material fed into them. Alpha calutron tanks were the larger of the two, and were often arranged in oval groupings called racetracks. Ninety-six tanks, alternating with giant electromagnets, stretched 122 feet in length, 77 feet wide, and stood around 15 feet high.

Tubealloy entered the calutrons as a salt, TC14: unremarkable brownish-greenish crystals. A heater cranked up the temperature until that Tubealloy salt vaporized. An electron filament producing highly charged electrons zapped the vaporized Tubealloy, ionizing the atoms. Now, the Tubealloy had a positive “charge.”

Pump a charged ion through a magnetic field and its path will bend, with the radius of that path dependent on its
mass.
So charged Tubealloy ions traveling through a magnetic field traveled along a semicircular path, with the heavier T-238 traveling on a radius larger than that of its lighter counterpart, T-235. At the end of this magnetized trip was a collector with two delivery slots, targets for the slightly different paths of the 238 and the 235. The Tubealloy ions slammed into a metal plate, looking like tiny metal flakes. The 238 was captured in one receiver—the Q slot—and the precious 235 was collected in another—the R slot. The distance between the slots at the end of the Tubealloy’s magnetically charged journey was roughly
³
/
₁₀
inch.

Calutrons were not called as such by the vast majority of workers at CEW, but were more commonly referred to as D units because when viewed individually they looked like giant letter Ds. Cubicle operators sat in front of giant stations working knobs and levers that controlled source heating, voltage, and ionization by visually monitoring their panels.

Beta calutrons were roughly half the size of the Alphas and were arranged in a rectangular configuration. T-235 collected from the Alpha calutrons was fed into the Beta calutrons for a second stage of enrichment. After a run through the Alpha calutrons, a batch of Tubealloy was enriched to about 12 or 15 percent 235—not a high enough percentage of 235 for the Gadget. After a Beta run, the enrichment level reached around 90 percent T-235, which
was
good enough for the Gadget.

The gist was the same for both Alpha and Beta runs: Tubealloy went into the calutrons as a salt, was ionized and accelerated through a magnetic field, and finally came out of the calutrons separated into two different isotopes: T-235 and T-238.

Workers removed Tubealloy from the collection boxes and scrubbed out the rest of the unit where possible with nitric acid. Everything was processed, to retrieve as much of the Tubealloy as possible. Workers’ clothing was often processed as well, to retrieve even the smallest bit of usable material. Before, after, and between Alpha and Beta runs, Tubealloy passed through hands and beakers and spectrometers and centrifuges and dry boxes. As it assumed its different forms, it was given a variety of code names: 723 for TO3, a yellowish powder, 745 for TCL5, greencake, yellowcake, and so on. A slew of chemists worked with the Tubealloy, like chefs hovering over a secret ingredient of a recipe they were not permitted to know. And at every turn, in every department, individuals accounted for the Tubealloy, from building to building, lab to calutron. Any time even a dusting of Tubealloy was transferred, someone filled out a waybill. How much. Assay. Analysis. Code number. Clerks hand-cranked calculators, couriers ferried sealed envelopes from building to building.

When completed, K-25 would be the largest building in the world, but the scale of Y-12’s operation was equally astounding. Magnets, for example, commonly require copper. And Y-12’s magnets were massive—the Alpha magnets were more than eight feet tall. But the rest of the war—the not-so-secret parts of the war—needed copper, too, for things like shell casings. So the Project used silver to construct its magnets. Who had a few tons of spare silver lying around? The US Treasury.

When Y-12 was being built, the District Engineer met with Under Secretary of the Treasury Daniel Bell to discreetly request around 6,000 tons of silver. This baffled a man normally accustomed to speaking of silver in troy ounces. A more official request followed, from the secretary of war to the secretary of the treasury: They agreed not to discuss specific details about how the silver would be used, only that it would, at some point, be returned to the federal government. Eventually, about 13,540 short tons of silver (395 million troy ounces) were borrowed from the US Treasury, taken from government vaults at West Point, and used to construct the calutron magnets. Price tag: more than $300 million.

While K-25 was still being constructed, Y-12 was already a complex of structures: Alpha and Beta calutron buildings, cooling towers, chemical processing, change houses, pump houses, steam plants, cafeterias, and much more. Buildings sometimes had various designations. Several of the buildings, despite the secrecy surrounding the Project, began with the number 92: Tubealloy’s atomic number. When the General first learned of this numbering trend, he did not find the choice particularly amusing or wise. He would later say the plant names were essentially random: the X in X-10 probably came from Site X. Y-12’s moniker was meaningless. The K in K-25 was for Kellex or Kellogg, the company responsible for that plant’s design and development, and 25 was used “throughout the Project,” according to the General, to refer to T-235.

HOW BIG IS BIG ENOUGH?

Tennessee Eastman, a division of Eastman Kodak, oversaw operations at Y-12, and though the company was vigorously recruiting for Y-12, there never seemed to be enough workers.

At first, the General and his team estimated that 2,500 people should be able to operate Y-12, but as early as the fall of 1943, nearly 5,000 workers had been hired. On September 9, 1943, the General had ordered the size of Y-12 to be doubled, in response to yet another revised estimate regarding the amount of T-235 needed for the Gadget.
More! Much more!
—had become an unfortunately familiar and unwelcome refrain. At that point, not even a year had passed since breaking ground at CEW when the Scientist and his Los Alamos team said they needed
three times
the amount of T-235 stated in their most recent estimate.