For many years, even centuries, atomic structure and then nuclear structure had been studied and poked at with tastefully modest experimentation. Everything had transpired as it should have, with experimentation and theory tossing the little badminton cock back and forth across the net. Everything theory could think up, experimentation could prove, and every time experimentation fell over something in the dimness, theory could shine some light. Everything in nuclear physics that was found, pondered, or anticipated was reported, first in the academic literature and then in the popular press. Nuclear physics was almost recreational in its sweep of the professional scientific and popular cultures.
Everything changed with World War II. Nuclear physics instantly became a military secret. All journal publications stopped. Physics had to leave the open, free-thinking, cross-pollinating atmosphere of universities and be relocated to dungeon-like venues. Physics departments in the schools were swept of personnel, carted off to secret laboratories built in out-of-the-way places, and sworn not to divulge anything. Immense working complexes were built in a rush, having names like “Site W” and “Site Y.” Constructors never knew what they were building. Industrial laborers working at the labs were not told what they were making, and they could not discuss it with the person sitting next to them, much less with anyone outside the fence. Those who had an inkling as to what was going on could not utter key words, such as “uranium-235.” Everything had a code-word. Uranium-235 was “oralloy” and natural uranium was “tuballoy.”94 Isotope separation plants were not known from their function, but from their locations, named Y-12, K-25, and S-50, on a secret map. Laboratories and production facilities were ringed with barbed wire, armed watch towers, and mine fields. Mail was censored, photography was forbidden, and any scrap of paper on which a scientist scribbled an equation became a controlled document. Robert Sproul, President of the University of California and administrator of a large nuclear lab in New Mexico, had no briefing as to what his laboratory was doing.95 Laura Fermi, wife of Enrico, did not know what her husband did for a living until days after the war ended.
Every odd sound, light flash, or truck moving into or out of a nuclear facility had a counter-story written to explain it to concerned citizens. There were many foreign scientists working on the project, and they often found themselves being tailed by the FBI if they left the laboratory grounds. Given the immense sweep of the atomic bomb project, involving thousands of workers, billions of dollars, and talkative physics professors, the security apparatus was superb. Nobody on the other side of the war had a clue as to what was going on. The Germans knew nothing about the project. The Japanese were unaware of it until the end product was dropped on them. Harry S Truman, Vice President of the United States and a heartbeat away from Commander in Chief, was kept completely uninformed until the death of FDR and his inauguration.
In this atmosphere of complete blackout, nuclear technology jumped forward. It made the sudden transition from a graduate textbook topic to a well developed industrial process in just three years, and its partial disclosure at the end of the war would be a shock to the outside world. It was a jolt in the dark.
Nobody outside knew what was going on, with the exception of the Soviets. Having practically invented modern espionage, it should be of no great surprise that they knew immediately what the United States was up to. The evidence was readily available and it was conclusive. It was not information that was written, said, or transmitted by radio. It was the sudden absence of information. The American physics journals had abruptly stopped publishing papers concerning nuclear topics, and there was only one explanation.
Georgii Flerov, one of a scant few Soviet nuclear scientists, pointed this out to Joseph Stalin, General Secretary of the Communist Party. Actions were soon set in motion in two directions. An atomic bomb project was started, to no great effect, with Igor Kurchatov the scientific head of the effort. All nuclear physics work was relocated to the village of Sarov.96 Also an infiltration of the American bomb project was initiated. The success of this spy action would be legendary, and it would save the Soviet Union a lot of redundant work after the war when they developed their version of the atomic bomb.
By September 13, 1942, it was time to get serious about the A-bomb project, and the S-1 Executive Committee held a crucial meeting. Present were Lyman Briggs, the Committee Chairman; James Conant, President of Harvard University; Arthur Compton, physicist and Nobel laureate; Ernest Lawrence, the cyclotron master from Berkeley; Egar Murphree, Standard Oil petroleum chemist; and Harold Urey, another chemist, but with a Nobel Prize under his belt.97 Absolute secrecy was necessary, so a special place was chosen in which these notables could meet and not be noticed. The meeting was held at 20601 Bohemian Avenue, in Monte Rio, California, in the middle of a deep forest. It was the meetinghouse of a secret, exclusive, all male organization known as the Bohemian Club.98
Founded in 1872 by, of all things, newspaper journalists, the club is ensconced in an isolated grove of redwood trees over 1,500 years old on 2,700 acres, with a placid lake in the middle. Its working motto carved over the entrance, “Weaving Spiders Come Not Here,” would be ignored as the S-1 Committee interlaced plans for the intricate tasks of turning nuclear theory into a weapon.99
That report from the British MAUD Committee, on ice in Briggs’ safe since 1941, was finally opened, and it pointed out a significant finding. It was known that uranium would fission by neutron capture at very low, thermal speeds, but there was another neutron fission point, at the other end of the energy spectrum. Fast neutrons, freshly flung by the fission process and traveling at 1 MeV, could also initiate fission in U-235, and the fission would occur promptly, without a delay. This meant that bombs could be made small and light, because uncontrolled, explosive, chain-reacting fission could be initiated in uranium without the use of a bulky moderator assembly like the one in Chicago. Instead of dropping a graphite pile the size of a house, an airplane could drop a bomb made of pure U-235, the size of a pineapple. An atomic bomb suddenly started to look almost practical.
The men understood the implications of this report, but they also admitted that they were already in over their heads. In three days of work they established a new, centralized laboratory to do nothing but study the effects of fast neutrons, to be code-named “Project Y.” To head it they needed a military officer who could impart some discipline to a bunch of scatter-brained academics; a West Point Man; a career soldier with some crust on him. They chose Colonel Leslie Richard Groves of the United States Army Corps of Engineers, just coming off a long project where he built a large, 5-sided building in Virginia called “The Pentagon.”100
Groves was born in Albany, New York, in 1896, and had a fine education at the University of Washington, the Massachusetts Institute of Technology, and West Point, where he was fourth in the class of 1918. He was worn out from the massive construction project and looking for a vacation. Instead, four days after the end of the S-1 meeting, he was assigned the duty to run this new project and was immediately promoted to Brigadier General.
The project was designated “Laboratory for the Development of Substitute Materials,” in the tradition of giving secret operations confusing names. Groves hated it. He changed it to the name of a nonexistent office of the Corps of Engineers, “Manhattan Engineering District.” While that was the official title, to everyone but Groves it would simply be “The Manhattan Project.” The budget was essentially unlimited, and he could have whatever or whomever he needed to complete the job. The singular task was to build and test something that still existed only in science fiction but not for much longer: the atomic bomb. Seven days after his appointment, his advance men bought a 52,000-acre tract of land in rural Tennessee, hidden between two mountain ranges, called Oak Ridge. Sparsely populated, forgotten by civilization, but recently endowed with a severe excess of electrical power by the Tennessee Valley Authority, it was perfect. The natives were swept out; temporary, pre-fabricated housing was installed; and a city was built where no Germans or Japanese would think to look for one. It would never be subjected to bombing, surveillance, or sabotage. Some called it Site X, some called it the Clinton Works, but the Manhattan Project called it Headquarters. Nobody dared call it Oak Ridge, for fear of a stern security lecture.
Under General Groves’ steady guidance, rapid progress was made. Decisions and directives that once took months to be implemented now took minutes to come to pass. Budgets that were in the thousands of dollars were now in billions. Even with all his newfound power, Groves needed help managing a hopeless force of Nobel laureates, foreign-born scientists, physics lecturers, and a gathering swarm of experts in disciplines that had yet to exist. He needed a top scientist to assist him, a theoretician who could manage a level stare at a quantum mechanic. He chose Dr. Julius Robert Oppenheimer.101
Oppenheimer was a theoretical physicist, educated at Harvard, Cambridge, and Göttingen, professing physics at Berkeley and working on the fast neutron study for the S-1 Committee. A slight snag was his security clearance. Oppenheimer had socialist tendencies, and he subscribed to People’s World magazine. His wife, his mistress, his landlady, his brother, brother’s wife, and at least five of his students were card-carrying members of the Communist Party, and it was not the type of party where you get silly on Friday night. It was the other form of party, the kind that could potentially plot the overthrow of the United States Government. Nothing stood in the way of Grove’s plans, however, not even the FBI, and Oppenheimer’s clearance breezed through.
Groves was a heavy fellow, and seemed to wear clothing that was a couple of sizes too small. He was addicted to Turtles candies, a confection made of pecan fragments in nougat, covered with milk chocolate. He was a straight military officer, trained in engineering, who could get on the phone, find a thousand yards of Readymix concrete, and have it delivered the next morning. He was direct and somewhat brusque, and his politics were conservative. He had read the World Almanac, from cover to cover.
Oppenheimer appeared to be starving, and his clothing seemed stored temporarily on his body, as he chain-smoked and drank. He was a career academician, trained in theoretical physics, who could bring a student to tears over an abstruse topic of mathematics, was capable of impressive genius, and once tried to strangle a friend over frustration with experimentalism. He could project arrogance and sarcasm, and his politics leaned left even at Berkeley. He read the Bhagavad Gita in the original Sanskrit.102
They were a perfect team. Both were driven by the need for success, which they achieved beyond expectation, and both could rightfully claim credit, or blame, for the rapid development of nuclear technology under war footing. They had to admire each other, and they became fast friends.
Groves had picked out his secret lab site in Tennessee, and Oppenheimer immediately picked out his in New Mexico. “Site Y” was the Los Alamos Ranch School, a boy’s preparatory academy located on an isolated mesa, surrounded by federally owned land, and having easily secured entrances. Oppenheimer had spent summers there as a boy, riding horses and generally roughing it. The project bought the entire school and everything in it, including harnesses for the horses, and the Oppenheimer family moved into the school administrator’s residence. All the existing buildings were put to use, and dozens of others were thrown together. As a place to experiment with explosives and dangerous materials, it was perfect.
By the end of 1942, with Fermi’s reactor experiment running full time in Chicago, Groves’ headquarters picked out, and ground broken at Oppenheimer’s Los Alamos Lab, a rough design of this new nuclear weapon was in the firm concept stage. Its explosive character would depend on three properties of nuclear fission that were not necessary for simple power production: Hyper-criticality, prompt fission, and fast fission.
Simple criticality was the signature of Fermi’s CP-1 nuclear power reactor design. Fission in a power-producing reactor occurs on a steady-state basis, with the mass of uranium kept to a size that ensures that for every neutron lost in a fission event, a new one is released.103 The rate of power production neither rises nor falls, as the fission reaction sustains itself over time by producing exactly as many usable neutrons as are lost in the process. A well-balanced reactor is said to have achieved criticality. With more uranium than is necessary to maintain a critical balance, the reactor goes into super-criticality, and in this state the power level rises exponentially. A reactor with a slight amount of extra fuel, say one percent, is super critical. Include enough excess fuel to make it three or four hundred percent over critical, and the reaction is hyper-critical. Power increases explosively.
In a nuclear reactor producing power at a steady state, the neutron production is exactly balanced, meaning that all neutrons produced are accounted for. Most of the neutrons produced in fission are released from the disintegrating uranium nucleus immediately, or instantly. A percentage are delayed, and the delay can range from a fraction of a second to several seconds. This is fine for a power reactor. It does not really matter when a neutron is produced, as long as it makes its way into the final accounting, but in a bomb everything has to happen in microseconds. Only the neutrons produced promptly in the reaction can be counted on for the chain reaction. A prompt critical mass is larger than a delayed critical mass. A hyper critical bomb mass is slightly larger than a hyper critical reactor mass, because of the accounting for the delayed neutrons which cannot participate in the explosion. By the time the delayed neutrons show up, everything will be gone.
In Fermi’s power reactor, the fissions were initiated by neutrons slowed down to crawling speed by the use of a graphite moderator. The reactor used natural uranium fuel, only weakly infused with fissile U-235, and every advantage had to be exploited, including the unusually large probability of fission at the very low neutron speed. This was fine for a constantly running, well controlled reactor, but for an explosive device this would not work. Too much time was wasted slowing neutrons down, and the reaction was too diffuse. It was possible to initiate fission using neutrons traveling at high speed, fresh out of the last fission, but it would take a larger mass of fuel to be critical at the disadvantageous high speed, and the fuel would have to be pure, with no worthless atoms standing in the way. Fast fission, it is called.
Put these factors together, and you’ve got a bomb. The Army Corps of Engineers formulated a plan to turn theory into weaponized hardware, and it was brilliant. The development would proceed on two independent, parallel paths, each with non-intersecting difficulties. The success of only one path was necessary, and in this way, they doubled their odds.
Path 1 was to build a bomb using uranium. Sufficient uranium ore was acquired, and the only thing standing between the Army and a uranium-based bomb was the fact that it had to be enriched, throwing away 99.3% of the material and saving the last 0.7%, which was the pure U-235. This would be difficult. There is no chemical reaction that can separate two isotopes of the same element from each other. The only difference between the dominant U-238 and the minority U-235, aside from the fissile nature, is the weight of an individual atom. The difference is not great. One weighs 238 AMU, and one weighs 235 AMU.104 A few methods of isotope separation, including the ultra-centrifuge and the mass spectrometer, had been tried in laboratory settings, managing to separate countable numbers of atoms. This would have to be isotope separation on a heavy industry level, resulting in kilograms of pure U-235 metal, and not merely detectable traces of metal. Isotope separation became the priority function of the laboratories at Oak Ridge, the Clinton Works.
Success of an isotope separation was risky, so this program was further divided into two parallel paths. Only one had to work to make the uranium bomb of path 1 successful. Path 1A was to use brute force, always a route worthy of discussion in industrial matters. Ernest Lawrence from Berkeley, of cyclotron fame and Nobel Prize, proposed that they should make a lot of very large devices based on the mass spectrometer principle.105 Lawrence was at the top of his game at this time. He was confident to the point of cockiness, persuasive, blond, with a California tan, and he was certain that he had a workable plan. The plan did work, but just barely. Uranium metal would be broken into individual ions under a hard vacuum and accelerated to speed by an electrical charge, aimed to hit a target. The path of the speeding ions would be bent into a tight turn using a constant magnetic field cutting through the vacuum. With all the uranium ions traveling at the same speed, the lighter ones, or the U-235, would take a tighter turn than the heavier U-238, and would therefore impact the target at a slightly different angle. Lawrence managed to find cracks in Groves’ crusty, larger-than-life personality and slip his bold proposal past the protective coating. It took a large imagination to see it work, but Groves started to believe, and he gave the plan the green light.
Lawrence named his separation machines “calutrons,” installed in utter secrecy at Oak Ridge, at the map coordinates Y-12. Only two stages of separation were needed, stage alpha and stage beta. Five alpha machines, called “racetracks” for their oval shape, were eventually built, giving a product with the U-235 enrichment boosted to 12%. Each racetrack, 122 feet long and 77 feet wide, contained 96 calutrons sharing a common magnetic field. The magnetism was created electrically, and for this the large generating capacity of the newly installed Tennessee Valley Authority was used to its fullest capacity.
Maintaining a constant magnetic field was important to make sure that the U-235 hit its collection target, but it was tricky to accomplish. Rooms full of operators, mostly young women, sat on stools, watching a voltage dial and adjusting a knob to keep the needle on the dial pointed straight up, 24 hours a day, correcting for the slight voltage variations that occur in a power network. It was code-named the “Z regulator.” It worked. The operators, recruited from the nearby Appalachian hill country, did not have a clue as to what they were doing or why.
A problem with building the alpha calutrons was a shortage of copper for magnet windings due to the war effort. The shortage was so acute that the U.S. Mint was currently making pennies out of steel. Over 12 tons of metal was needed for the massive electromagnets. Never meeting a problem he didn’t like, General Groves rolled over it without losing speed, like a Sherman tank flattening a chicken house. If they could not get copper, then silver would do just as well, if not better. Groves borrowed all the silver needed for Lawrence’s separators from the United States Treasury Department, $300 million of it, kept in storage at the West Point Depository.106
There were many problems with the calutron setup, as this was a completely new technology and there had not been time to build and test a pilot plant. The first unit failed from contaminated cooling oil, and the second was barely limping along by January 1944. It became evident that the second stage would be necessary, and a set of smaller, more precise beta calutrons was designed and installed at Y-12. By early 1945 there were six beta machines running with a 12-percent feed from the alpha units, turning out 80-percent enriched U-235. It was a terribly wasteful process. Less than 5% of the precious U-235 managed to hit the collection target, with the rest splattered over the inside of the calutron and remixed with the U-238. They needed about 10 kilograms of it to make one bomb, and it seemed an excruciating effort to collect it.
Meanwhile, Path 1B for uranium enrichment was under parallel development at map coordinates K-25, eight miles southwest of Y-12. The second separation method was gaseous diffusion, as promoted and developed by Dr. Harold Clayton Urey at Columbia University. Urey had won the Nobel Prize in chemistry in 1934 for the discovery of deuterium, and he was a world-class expert on isotope separation. The gaseous diffusion method was locked down under military secrecy, large-scale machinery was designed, and construction was begun at Oak Ridge in June 1943.107 The entire plant was under one roof, half a mile long by 1,000 feet wide, with a floor area over 2,000,000 square feet. In 1943 it was the largest building in the world.
The gaseous diffusion method of isotope separation exploited the fact that a lighter molecule of gas will diffuse slightly faster through a permeable membrane than will a heavier molecule. It was a simple principle, but the effect was meager, and the process would have to be repeated thousands of times to produce measurable results. There were also a few practical problems. Diffusion works with gas, and only gas. Uranium is a solid metal. The way to make a uranium gas at standard temperature is to produce the compound uranium hexafluoride. The fluorine component is composed of a single isotope, so it does not confuse the issue of molecular weight. Unfortunately, uranium hexafluoride is extremely reactive, and most materials that come into contact with it are dissolved. Nickel, however, does not dissolve in uranium hexafluoride. Everything in the building—pipes, flanges, valves, seals, pumps, diffusion chambers—was made of pure nickel or nickel-loaded austenitic stainless steel. From an industrial perspective, the sight of the inside of the K-25 building, with its clean and orderly maze of gleaming nickel plumbing seeming to extend forever and disappearing into the haze at semi-infinite distance, was beautiful.
It took 15,000 people to erect and operate the gaseous diffusion plant at K-25. All employed by the Army, they lived in an immense trailer-park assembled near by, given the fond euphemism “Happy Valley.” By the summer of 1944, the population of the Headquarters of the Manhattan Project had grown to 50,000 people, only a few of whom knew exactly what was going on. It was the only place in the world where a disgruntled employee could whack you over the head with a pipe made of pure nickel.
The K-25 plant, for all the expense of $512 million dollars, for all the ingenious concept and the incredible accomplishment of having built it in secrecy, did not reach operating capacity during the war. Diffusion, by its very nature, takes time, and to achieve full separation from neighboring U-238 one U-235 hexafluoride molecule has to diffuse through thousands of cascading steps in a string of diffusion chambers over a mile long. By the time the first high-purity U-235 was available at the end of the cascade, the war was long over. However, the plant was hardly a worthless exercise. Compared to the Y-12 plant, gaseous diffusion was an efficient and non-wasteful process. At the end of World War II the United States had the only uranium enrichment plant in the world, turning out U-235 in any desired concentration in quantities limited only by the amount of material loaded into the front-end.108 The availability of this one facility would influence the path of nuclear power development in the United States, make it different from the course taken in any other country with nuclear ambitions, and permanently affect worldwide reactor design.
Soon after the end of the war, the valuable magnet windings of the calutrons in Y-12 were dismantled and quietly returned to the Treasury. Yet a third separation plant, the steam-powered thermal column cascade located at map coordinates S-50, was also broken down after the war. It was a method used by the Navy in its own secret program to develop a nuclear submarine engine, hidden away in the Philadelphia Navy Yard at the beginning of the war. Upon learning of its existence, Groves appropriated the equipment and gave the contractor, H. K. Ferguson, 90 days to build a thermal column plant at Oak Ridge. It was a weak process, and could only give 0.15% enrichment, but the Manhattan Project was desperate, and any fraction of a percent was better than nothing.
In the last weeks before the bomb was assembled Oppenheimer was growing frantic over the lack of enriched uranium, and he ordered that the three processes be lined up in series. The S-50 took raw uranium, increased it to 0.85%, and then fed the result to K-25. K-25 took it up to 20% and fed the result to the alpha calutrons, and from there to the beta calutrons. Oppenheimer thus got his 82-percent U-235, with days to spare.
Path 2 of the atomic bomb development would not depend on isotope separation. The man-made fissile isotope of plutonium, Pu-239, can be produced in industrial quantities using a scaled-up version of Fermi’s CP-1 graphite pile from the experiment in Chicago. Simply running a nuclear reactor at high power produces Pu-239, as the inert U-238 in the fuel captures stray neutrons and transforms to neptunium-239, which decays upward into plutonium. The CP-1 pile remained assembled for 90 days, and in that time enough was learned from experiments to design the first nuclear power reactor.109 The pile was then disassembled and rebuilt at Red Gate Woods, a forest preserve near Chicago. The pile was renamed CP-2, and research continued.
A scaled-up version of the CP-1 was designed for Site W, a spot picked out in the middle of the flat desert in Washington State, near Richland.110 It was named the Hanford Works, and like the other secret laboratories it was built in complete privacy, far from curious eyes and far enough from civilization that a major nuclear disaster on site could harm only those working there.
CP-1 produced about a half watt of power. The new plutonium production reactor produced 250 megawatts, enough power in the current century to supply 180,000 people with electricity. The purpose of the new graphite reactor was not, however, to make electricity, it was simply to make plutonium, and all its produced power would be wasted into the Columbia River, running through the sprawling property. The first machine, named B-reactor, was housed in a building the size of a gymnasium, and it used pure graphite bricks as the neutron moderator. Natural uranium was formed into slugs, each about the size of a roll of quarters, sealed in an aluminum can. Fuel slugs were lined up in aluminum tubes, running horizontally through a graphite cylinder, 28 feet around by 36 feet long, lying on its side. The graphite alone weighed 1,200 tons, and there were 1,500 tubes of fuel. Water was sucked out of the river and force-pumped through the fuel tubes at 30,000 gallons per minute to remove the energy and keep the thing from melting.
The Manhattan Project was serious about making plutonium and avoiding the tedious uranium separation. The plant was half the size of Rhode Island, with 512 specialized buildings and cities of temporary housing. Construction involved 42,400 workers, and after two intense years of work, on Tuesday evening, September 26, 1944, the B-Reactor was awaiting startup.
The Hanford Works was far more than a power reactor. It would eventually be a collection of reactors, lined up on the river-bank, but it was mainly a chemical plant, designed to extract the newly created plutonium from spent reactor fuel and form it into pure metal. E. I. du Pont de Nemours and Company did the engineering, and they were contracted to run the complex. Enrico Fermi, Pope of Los Alamos and Mother Superior of the pile technology, was there to witness the fuel being loaded and confer his blessings. The top executives of du Pont, perhaps not completely understanding what they were seeing, were on hand as well to experience the historic launching of the mysterious process. Everything checked out perfectly, and the thoroughly trained operation staff pulled the controls in small stages, just as Fermi had done in Chicago.
By about midnight, the pile had achieved criticality, and was sustaining the chain reaction of controlled fission. With due caution the power was increased. The pumps came online, and by 2:00 A.M. the reactor was operating smoothly at high power, dumping hot water into the river. Everyone was pleased, and congratulations were exchanged all around.
The good times rolled for about an hour, and then things took a downward turn. To maintain power, the operators had to start stepping out the control rods. They ran out of rods to pull, and early Wednesday morning the B-Reactor rolled over and died in its sleep. The power level gently decreased to zero, and no self-sustaining fission could be coaxed out of the dead machine. Fermi found this puzzling. He managed to remain calm though nervous as he reviewed, over and over, what the operators had done. Somebody had to call Groves and tell him the news. Groves, constantly juggling multiple crash programs, was tightly wound, and on this project he had riding several hundred million taxpayer’s dollars and the entire Plan 2. Fortunately he was in a distant location and could not reach through the phone and strangle anyone. He was not pleased, and neither was du Pont.
John A. Wheeler, a noted nuclear theoretician from Princeton University, was also present at the crisis, and he noticed something. With the time it took for the reactor to shut down, there was obviously some fission product that was decaying with a fairly short half-life into something that gobbled up neutrons. With a reduced neutron count, there were no self-sustaining nuclear reactions. His guess was iodine-135, decaying with a half-life of 6.68 hours. There was nothing inherently wrong with I-135, and it was a known fission product, but I-135 decays into xenon-135. That particular isotope of xenon is a voracious neutron absorber, 150 times more effective at eating neutrons than the cadmium used in the reactor control rods. The team had just observed the effects of what would come to be known in the annals of nuclear engineering as “xenon poisoning.” At low power, there is no problem, as the noble gas xenon will not stick to anything in the structure and is free to wander out of the reactor and into the surrounding air. It seeps out faster than it is made. No one had noticed it at the low powers used in the test reactors. At high power the rate of all fission-product production is much faster. The xenon builds up in the fuel and kills the process.
Wheeler had anticipated a problem, and had insisted that du Pont build the production reactors with more fuel channels built into the graphite than was necessary. He could not put his finger on the exact problem, but he could feel something coming. Du Pont howled, as the extra capacity added millions of dollars to the cost per pile, but Wheeler had insisted, with no specific reason. The extra fuel channels would allow 504 more lines of fuel slugs, and this was enough excess reactivity to overcome the xenon poisoning effect. The next two reactors in the series, D-Reactor and F-Reactor, were started in December 1944 and February 1945, using the enhanced loading of 2,004 tubes, and B-Reactor was re-configured. By April 1945 plutonium was flowing through the chemical process at Hanford and being shipped to Los Alamos. Plan 2 was in motion.
At Site Y, the Los Alamos Ranch School, Oppenheimer had organized an impressive group of nuclear scientists to take the puzzle to its conclusion. He got everyone he wanted from every top university. He had Enrico Fermi, Hans Bethe, Edward Teller, Stanislaw Ulam, Seth Neddermeyer, George Kistiowsky, and a host of others. He collected an impressive staff of technicians and specialists, experts in chemistry, explosives, precision machine-work, mathematics, and electronics. He had everything but Leó Szilárd, birth-mother of the whole thing. Szilárd was not acclimatized to roughing it or camping, preferring a hotel lobby and a magazine to the wide, open skies of New Mexico and the rustic living quarters. He balked at the suggestion that he should move out there and be where the action is. “Nobody could think straight in a place like that,” he protested. “Everybody who goes there will go crazy.” Also unconvinced was Isador I. Rabi, another very capable theorist from Hungary, who would not go but for different reasons. He was working at MIT on the secret radar development. “I’m very serious about this war,” he told Oppenheimer. “We could lose it with insufficient radar.” He was correct in that assessment, and Oppenheimer understood his priorities.
Still Oppenheimer continued to plunder, and he relieved Harvard of its cyclotron and the University of Wisconsin of two Van de Graff linear particle accelerators, to use them as neutron generators at his New Mexico lab. Otherwise, Oppenheimer seemed to impart little methodology to the construction project in Los Alamos. Tight security complicated things, but the entire security staff was composed of Mexican laborers, and few could speak English. The ranking Army officer in charge of the security office was a lieutenant. Oppenheimer engaged in a mini-war with the Army Corps of Engineers to prevent them from cutting down all the trees. It was like watching an ant farm get organized.111 Pick out an individual ant, and it seems to wander about, aimlessly picking up a particle of dirt here and there and dropping it somewhere. Come back the next morning, and an underground city has been built, with interconnecting tunnels. Somehow, the Los Alamos facility got built, almost overnight. Laboratories, shops, and offices sprang out of the desert, and in April 1943 his new facility opened with a school set up to indoctrinate incoming scientists and brief them as to what they were supposed to accomplish and what had happened so far. Secrecy covered everything like an itchy wool blanket, and the war invaded all thoughts, but Oppenheimer strove to prevent security and politics from stifling the much-needed creativity.
The short course lasted five days, taught by a young man from the University of Chicago named Robert Serber. They met on the first floor of the Technical Area library, with fifty people in folding chairs and Serber scratching away on a small blackboard set up in the front of the room. Carpenters and electricians were working all over the place, with hammering and sawing breaking concentration from all directions.112 Abruptly, at a particularly dramatic point in Serber’s exposition of the explosive character of fissile materials, a leg crashed through the beaverboard in the ceiling. A workman upstairs had misstepped. An air of spontaneity seemed to prevail.
The apex of the pyramid of tasks and divisions at Los Alamos was, at least at the first, the Theoretical Division, including a scientist from Germany, Hans Bethe. Bethe was from Strassburg and had earned his Ph.D. in physics at the University of Munich. He left Germany under pressure in 1933, worked in England at the University of Bristol, and then immigrated to the United States in 1935 to take a position at Cornell University in Ithaca, New York. Oppenheimer met him at Berkeley in 1942, in a special session studying the theory of the atomic bomb. Unlike most of the participants, Bethe thought that fission could not work explosively, and he had written a paper to prove it. Oppenheimer was impressed. He made him head of the Theoretical Division. Bethe changed his mind about fission and subsequently calculated the critical mass of U-235 required to explode.
Edward Teller felt passed over for the position of Theoretical Division head, but he immersed himself in the work and became the idea man, the one who kept big concepts coming on the tip of the spear. Unfortunately, most of his ideas were too far ahead, usually ten years from anything that they could conceivably build or even need. He kept the blackboard in his office loaded with a list of bomb concepts. Columns listed the properties of each bomb design, its destructive abilities, and finally, in the last column was listed the bomb delivery method. At the bottom of the board was his most ambitious weapon, and its odd delivery method was “back yard.” Since it would probably kill every living thing on earth, there was no use delivering it anywhere, and it might as well be exploded in the back yard.
With the other theorists operating on a more practical level, a bomb design was roughed out. As Serber explained in his short course, the mechanical problem was to bring together sub-critical masses of the fissile material, quickly, into a multiple of critical masses.113 How quickly? The masses had to assemble with haste, so hat the runaway reaction would not consume the fissile material before it reached its full explosive potential. The first bomb design used a cannon barrel to assemble two sub-critical masses of fissile metal into one exploding mass. It was an attractive idea for its simplicity. A cylinder of plutonium or uranium is accelerated to high speed by an exploding guncotton charge at the breech end of the barrel. By the time it reaches the end of the barrel, the cylinder is traveling at 3,000 feet per second. The sub-critical cylinder passes through a hollow cylinder of fissile metal, also sub-critical. Together the projectile and the target constitute a hyper-critical mass.114 The end of the barrel was kept open, as if the projectile could shoot out the end. It would never get that far. As the projectile approaches full contact with the hollow cylinder, the entire assembly fissions into a huge ball of super-hot plasma.
To build up enough speed on the projectile, the barrel had to be 18 feet long. It looked like a telephone pole with fins on one end and a two-foot-diameter bulge on the other end. Serber named it “Thin Man,” and by August 1943 the Navy had some models to drop-test, just to see how the thing would fly.115 The test bombs were not good, with each example going into a flat spin after drop and splattering to pieces when it pancaked into the ground. For all its virtues, aerodynamically the design was terrible.
The flight characteristic turned out to be the least of the problems. In July 1944, just as everyone started to think that the puzzle-phase was over, things began to get interesting. As larger samples of plutonium became available from steady production at Hanford, experiments were performed to get better data on its nuclear characteristics, and it became apparent that the man-made Pu-239 contains traces of Pu-240. The Pu-240 contaminant is a hair-trigger fissile isotope, and its tendency to pre-detonate would tear the cannon-barrel bomb to pieces before the projectile had a chance to mate with the hollow cylinder. The plutonium-based bomb would be a fizzle. On July 11 Oppenheimer broke the news to Conant at Harvard.
On a brighter note, tests on incoming samples of enriched U-235 proved that the pre-detonation tendency of urnaium measured out less than they had thought. For uranium, the Thin Man could be shortened to 10 feet. Serber renamed it “Little Boy.” The shortened bomb casing flew better. After drop it smoothly pitched over and flew nose-down, hitting the ground in a perfectly vertical attitude, as a bomb is supposed to hit.
There was no way they could do an isotope separation on Pu-239 and Pu-240. The masses of the two isotopes were too close together, and it was all they could do to separate U-235 from U-238. It looked as though the hundreds of millions of dollars and the precious time spent on Plan 2 were down the drain. Crisis meetings were held. Teller suggested that they move on to his hydrogen-fusion bomb idea. No thanks, Teller. Anybody else got an idea?
A young physicist from Caltech in Pasadena, Seth Neddermeyer, had been tossing around a concept since April 1943. When assembling a hyper-critical mass of plutonium or uranium, it was always assumed that two separate, sub-critical masses would be joined together. What if another aspect of the critical assembly could be changed, other than the total mass? What if a single, sub-critical mass could be made hyper-critical by suddenly increasing its density? Yes, that could happen in theory, but how does one change the density of an incompressible mass of heavy metal?
Metal does seem incompressible, but it is incompressible only to a point. Hit with enough pressure, anything can be compressed. If a ball of plutonium the size of an orange, sub-critical and perfectly safe under normal conditions, could be compressed to a ball the size of a large marble, then it would explode as the conditions for fission improved, without the troublesome delay of bringing two separate pieces together. Neddermeyer’s concept was radical and unprecedented: use explosives to compress the metal.
Everyone was aware of the actions of a high-explosive charge. The explosion is initiated at the center of a mass of fast-burning chemical, and a spherical shock-wave travels outward, dissipating the energy and the forces as the ball of shock increases in size. Neddermeyer turned the process around. Instead of initiating at a point in the center of the material, an explosion could be started simultaneously at all points on a surface, outside the ball of explosive. The shock wave would proceed inward, toward the center instead of outward into space. Instead of growing larger and dissipating force, the ball of shock grows smaller, and concentrates the force. By the time it converges at the center, the shock wave commands enough force to warp the molecular structure of solid metal and cause a ball of plutonium at the center to achieve hyper-criticality and explode under nuclear fission. Neddermeyer called it implosion. It was brilliant. Oppenheimer immediately put him in charge of the testing and research.
Even the smartest ideas can be very difficult to implement. Neddermeyer was not an expert on explosives, and his initial attempts to shape-charge an implosion were unproductive. The best that could be said was that nobody got killed, as he tried to compress various metal shapes. Oppenheimer brought in a Russian explosives expert, George Kristiakowsky, to review the lack of progress. Kristiakowsky was a chemistry professor from Harvard, and he could deliver an excruciating lecture on the properties of explosives at the drop of a hat. Oppenheimer received his report in mid-June, 1944, read it, and promptly made Kritiakowsky head of the implosion group, over Neddermeyer. Johnny von Neumann, another genius from Hungary, was brought in from Princeton as a consultant, to work through the difficult mathematical models of the shaped charges for the controlled implosion. The development of the implosion concept proceeded quickly.
The design of the plutonium-based bomb began to focus. It would not be a long, thin object. It would be a sphere of explosive, about 9 feet in diameter. Put some fins on the back of it to make it fly nose-down, and you have a very clumsy-looking bomb with a predicted performance that would be better than the simpler uranium bomb. Serber named it “Fat Man.”116 Both paths to the development of an atomic bomb, uranium and plutonium, or 1 and 2, converged on success.