Boris Pash in the meantime continued to chase down the German atomic scientists. Alsos documents placed Werner Heisenberg, Otto Hahn, Carl von Weizsacker, Max von Laue and the others in their organization in the Black Forest region of southwestern Germany in the resort town of Haigerloch. By late April the German front had broken and the French were moving ahead. Pash and his forces, which now included a battalion of combat engineers, got word in the middle of the night and raced around Stuttgart in their jeeps and trucks and armored cars to beat the French to Haigerloch. They drew German fire along the way and returned it. In the meantime Lansdale in London reassembled his British-American team and flew over to follow Pash in. The story is properly Pash's:
Haigerloch is a small, picturesque town straddling the Eyach River. As we approached it, pillowcases, sheets, towels and other white articles attached to flagpoles, broomsticks and window shutters flew the message of surrender.
… While our engineer friends were busy consolidating the first Alsos-directed seizure of an enemy town, [Pash's men] led teams in a rapid operation to locate Nazi research facilities. They soon found an ingenious set-up that gave almost complete protection from aerial observation and bombardment — a church atop a cliff.
Hurrying to the scene, I saw a box-like concrete entrance to a cave in the side of an 80-foot cliff towering above the lower level of the town. The heavy steel door was padlocked. A paper stuck on the door indicated the manager's identity.
… When the manager was brought to me, he tried to convince me that he was only an accountant. When he hesitated at my command to unlock the door, I said: “Beatson, shoot the lock off. If he gets in the way, shoot him.”
The manager opened the door.
… In the main chamber was a concrete pit about ten feet in diameter. Within the pit hung a heavy metal shield covering the top of a thick metal cylinder. The latter contained a pot-shaped vessel, also of heavy metal, about four feet below the floor level. Atop the vessel was a metal frame… [A] German prisoner… confirmed the fact that we had captured the Nazi uranium “machine” as the Germans called it — actually an atomic pile.
Pash left Goudsmit and his several colleagues behind at Haigerloch on April 23 and rushed to nearby Hechingen. There he found the German scientists, all except Otto Hahn, whom he picked up in Tailfingen two days later, and Werner Heisenberg, whom he located with his family at a lake cottage in Bavaria.
The pile at Haigerloch had served for the KWI's final round of neutron-multiplication studies. One and a half tons of carefully husbanded Norsk-Hydro heavy water moderated it; its fuel consisted of 664 cubes of metallic uranium attached to 78 chains that hung down into the water from the metal “shield” Pash describes. With this elegant arrangement and a central neutron source the KWI team in March had achieved nearly sevenfold neutron multiplication; Heisenberg had calculated at the time that a 50 percent increase in the size of the reactor would produce a sustained chain reaction.
“The fact that the German atom bomb was not an immediate threat,” Boris Pash writes with justifiable pride, “was probably the most significant single piece of military intelligence developed throughout the war. Alone, that information was enough to justify Alsos.” But Alsos managed more: it prevented the Soviet Union from capturing the leading German atomic scientists and acquiring a significant volume of high-quality uranium ore. The Belgian ore confiscated at Toulouse was already being processed through the Oak Ridge calutrons for Little Boy.
At Los Alamos in late 1944 Otto Frisch, always resourceful at invention, proposed a daring program of experiments. Enriched uranium had begun arriving on the Hill from Oak Ridge. By compounding the metal with hydrogen-rich material to make uranium hydride it had become possible to approach an assembly of critical mass responsive to fast as well as slow neutrons. Frisch was leader of the Critical Assemblies group in G Division. Making a critical assembly involved stacking several dozen 1 ‘/i-inch bars of hydride one at a time and measuring the increased neutron activity as the cubical stack approached critical mass. Usually the small bars were stacked within a boxlike framework of larger machined bricks of berylHum tamper to reflect back neutrons and reduce the amount of uranium required. Dozens of these critical-assembly experiments had gone forward during 1944. “By successively lowering the hydrogen content of the material as more U235 became available,” the Los Alamos technical history points out, “experience was gained with faster and faster reactions.”
But it was impossible to assemble a complete critical mass by stacking bars; such an assembly would run away, kill its sponsors with radiation and melt down. Frisch nearly caused a runaway reaction one day by leaning too close to a naked assembly — he called it a Lady Godiva — that was just subcritical, allowing the hydrogen in his body to reflect back neutrons. “At that moment,” he remembers, “out of the corner of my eye I saw that the little red [monitoring] lamps had stopped flickering. They appeared to be glowing continuously. The flicker had speeded up so much that it could no longer be perceived.” Instantly Frisch swept his hand across the top of the assembly and knocked away some of the hydride bars. “The lamps slowed down again to a visible flicker.” In two seconds he had received by the generous standards of the wartime era a full day's permissible dose of radiation.
Despite that frightening experience, Frisch wanted to work with full critical masses to determine by experiment what Los Alamos had so far been able to determine only theoretically: how much uranium Little Boy would need. Hence his daring proposal:
The idea was that the compound of uranium-235, which by then had arrived on the site, enough to make an explosive device, should indeed be assembled to make one, but leaving a big hole so that the central portion was missing; that would allow enough neutrons to escape so that no chain reaction could develop. But the missing portion was to be made, ready to be dropped through the hole so that for a split second there was the condition for an atomic explosion, although only barely so.
Brilliant young Richard Feynman laughed when he heard Frisch's plan and named it: he said it would be like tickling the tail of a sleeping dragon. The Dragon experiment it became.
At a remote laboratory site in Omega Canyon that Fermi also used, Frisch's group built a ten-foot iron frame, the “guillotine,” that supported upright aluminum guides. The experimenters surrounded the guides at table level with blocks of uranium hydride. To the top of the guillotine they raised a hydride core slug about two by six inches in size. It would fall under the influence of gravity, accelerating at 32 feet per second/per second. When it passed between the blocks it would momentarily form a critical mass. Mixed with hydride, the U235 would react much more slowly than pure metal would react later in Little Boy. But the Dragon would stir, and its dangerous stirring would give Frisch a measure of the fit between theory and experiment:
It was as near as we could possibly go towards starting an atomic explosion without actually being blown up, and the results were most satisfactory. Everything happened exactly as it should. When the core was dropped through the hole we got a large burst of neutrons and a temperature rise of several degrees in that very short split second during which the chain reaction proceeded as a sort of stifled explosion. We worked under great pressure because the material had to be returned by a certain date to be made into metal… During those hectic weeks I worked about seventeen hours a day and slept from dawn till mid-morning.
The official Los Alamos history measures the significance of Frisch's Dragon-tickling:
These experiments gave direct evidence of an explosive chain reaction. They gave an energy production of up to twenty million watts, with a temperature rise in the hydride up to 2 °C per millisecond. The strongest burst obtained produced 1015 neutrons. The dragon is of historical importance. It was the first controlled nuclear reaction which was supercritical with prompt neutrons alone.
By April 1945 Oak Ridge had produced enough U235 to allow a near-critical assembly of pure metal without hydride dilution. The little bars arrived at the Omega site packed in small, heavy boxes everyone took pains to set well apart; unpacked and unwrapped, the metal shone silver in Frisch's workbench light. Gradually it oxidized, to blue and then to rich plum. Frisch had walked in the snow at Kungalv puzzling out the meaning of Otto Hahn's letters to his aunt; in the basement at Bohr's institute in Copenhagen he had borrowed a name from biology for the process that made these small exotic bars deadly beyond measure; at Birmingham with Rudolf Peierls he had toyed with a formula and had first seen clearly that no more plum-colored metal than now lay scattered on his workbench would make a bomb that would change the world. At Los Alamos in Southwestern spring, denouement: he would
