However, when Groves asked Oppenheimer to identify the three men whom Chevalier had approached, Oppie agreed to do so only on one condition: that Groves promise not to divulge the names to the FBI. Believing that the three were undoubtedly among Oppie’s four graduate students, whom the army already had under surveillance, Groves agreed.
But what he heard next surprised him. There had been only one person contacted by Chevalier, Oppenheimer said: his brother, Frank.
Filling out the story, Oppenheimer told Groves that several months earlier Chevalier had approached Frank about the possibility of either passing secrets on the bomb project to the Russians or persuading his brother to. Uncertain how to respond, Frank had come to Oppie for advice. His recommendation, Oppenheimer told Groves, was that his brother should have nothing further to do with the scheme—and he had later given Chevalier his “comeuppance” for trying to recruit Frank as a spy.77
On the flight back to Washington, Groves wondered whether Oppenheimer was telling him the truth—or whether Oppie had simply introduced Frank into the story as a way to justify his own earlier failure to notify the army of Chevalier’s overture. More important, Groves realized that by promising not to divulge Frank’s name to the FBI, he, too, was now unwittingly a party not only to a lie but to a felony—namely, withholding the truth about an espionage conspiracy from federal authorities in wartime.
Back at the New War Department Building, Groves called Lansdale and Army Major William Consodine, his chief troubleshooter and lawyer, into his office. Putting a yellow legal pad in front of them, Groves asked each man to write down his guesses as to the names that Oppenheimer had given. Lansdale wrote down three names, all presumably from Pash’s list. Consodine put down only one—Frank’s.78
When Groves revealed what Oppie had told him, the trio discussed whether the general should consider himself bound by his promise to Oppenheimer. As Groves’s lawyer, Consodine argued that the security of the nation represented a higher and overriding obligation.79 But Groves felt that the bomb project itself might be fatally compromised if he lost Oppenheimer’s trust by violating his pledge. When Groves raised the possibility of a secret prosecution of Chevalier and Eltenton, Consodine warned that not even the wartime emergency allowed the suspension of due process.80
Either on his own—or with Groves’s willing connivance—Lansdale decided that he would verbally inform the FBI of the information received from Oppenheimer, but without putting anything down in writing.81
Remarkably, Groves chose not to let his own former deputy, the MED district engineer, Kenneth Nichols, in on the secret.
On December 13, Nichols wired Lyall Johnson at Berkeley that Oppenheimer’s intermediary was Chevalier. But Nichols’s cable unknowingly perpetuated the earlier fiction that Oppie had told Pash: “Oppenheimer states in his opinion Chevalier engaged in no further activity other than three original attempts.”82 Nichols sent similar telegrams that same day to de Silva at Los Alamos and Calvert at Oak Ridge. Pash, who was already in Italy, working for Alsos, remained unaware of the latest developments.83
That same evening, Lansdale went to FBI headquarters to personally inform Hoover aides Lish Whitson and Frank Tamm that Haakon Chevalier had tried to recruit Robert Oppenheimer’s brother to spy for the Soviet Union.84
* * *
The first contingent of cyclotroneers arrived at “Dogpatch” in the early days of 1944 to find things in utter disarray.85 Delayed by another bad cold, Lawrence, with Cooksey in tow, detrained at Oak Ridge on January 20 to take control of the operation. Ernest was assigned a comparatively luxurious E-unit apartment on Tennessee Avenue, near the camp store and movie theater.86 With the help of a little long-distance troubleshooting by Brobeck, some of the teething problems with the Alpha racetrack had already been worked out by the time that Ernest appeared.87
Martin Kamen arrived in the second wave of Berkeley recruits, to supervise the process of chemically separating uranium, only to discover that he had left the package containing the necessary floats and filters behind on the train.88 Kamen subsequently found that even when the racetracks were working properly, only about 10 percent of the orange uranium oxide—the “feed” material—was converted by the Calutron beam into a focused, ionized arc. Of that, only a few percent emerged in the receivers at the other end. Most of the “gunk” was spewed by the beam around the innards of the Calutron. The vacuum tanks had to be dismantled at the end of each run and their contents laboriously scraped, dissolved, and precipitated out to be fed into the machine once more.89
Lawrence proposed ingenious ways to work around the remaining problems. When the Calutrons began to overheat, he had the Oak Ridge fire department play their hoses upon the cooling towers until temperatures dropped to a safe level. The design flaw was found and corrected. At Lawrence’s request, Frank Oppenheimer flew out to Y-12 late in the month to deal with a corrosion problem, caused when chlorine used in the chemical process reacted with nickel in the stainless steel pipe.90 Copper liners proved the solution and Frank returned to Berkeley.
At the end of January 1944, Alpha Calutron 2 became the first racetrack to begin continuous operation. The mistakes and problems that surfaced in the initial start-up meant that the original Alpha track would not be back on-line till March. Recently imposed security strictures added another headache. Henceforth, all personnel assigned to Y-12 from Berkeley had to be cleared first by the army.91
With the confidence born of experience, however, the pace began to quicken that spring. In March, the first U-235 was shipped to Los Alamos—a few dozen grams. It found immediate use in experiments at the lab.92 By the end of April, all four of the original
