The drawing I present here is from a Russian publication dated April 16, 1948.{17} The labels are in Russian, but the text is in English as well as Russian. It is widely assumed that this document and its drawing are taken from what Fuchs supplied to the Soviets in March 1948, but there is no way to be certain of this.{18} (There is still no English version published in the United States.) If Fuchs transmitted information about his joint invention with von Neumann to the Soviets in 1946, as he could have, it may have been more offhand, less detailed than in 1948. In any case, it seems not to have had a measurable effect on the Soviet program. Nor, for that matter, on the American program. I don’t recall it being discussed—although, as I shall explain in the next chapter, the Greenhouse George shot in May 1951 did have some features in common with what von Neumann and Fuchs had proposed.
For the “detonator” (the fission-bomb trigger) von Neumann and Fuchs proposed a gun-type weapon similar to the one dropped on Hiroshima. This is pictured on the left side of the drawing. A piece of uranium is fired to the right at another piece of uranium, creating a supercritical mass that explodes. Von Neumann and Fuchs were even quite specific in saying that in the two pieces there would be a total of 71 kilograms of U235. The exploding mass continues to slide to the right and slams into a capsule of DT (the “primer”) held within a spherical shell made of beryllium oxide (the circle on the right of the diagram). They envisioned an initial threefold compression of the DT as a result of this collision.
Then comes the radiation and the ingenious part of the von Neumann-Fuchs invention. The flood of radiation from the fission bomb would completely ionize the BeO shell and its DT contents—that is, strip all electrons from all the atoms, leaving a plasma consisting of electrons and atomic nuclei. Since beryllium is the fourth element in the periodic table and oxygen is the eighth, where there had been one Be atom and one O atom there would now be fourteen particles (four electrons and one nucleus from the Be atom, eight electrons and one nucleus from the O atom). Inside the capsule, however, one D atom and one T atom would together yield only four particles (two each). So the BeO plasma, now having many more particles per unit volume than the DT plasma, would experience a much greater increase in its pressure and would exert a mighty compressive force on the DT. Von Neumann and Fuchs estimated that this one-two punch of a direct hit by the fission bomb followed by a radiation-induced implosion of the BeO capsule would result in a ten-fold compression of the “primer”—enough, they concluded, to get thermonuclear ignition.
The burning “primer,” they then concluded, could ignite the “booster,” which, in turn, would set the “main charge” aflame, with multimegaton consequences. Again being quite specific, they suggested that the “main charge” would reside in a cylinder about two feet in diameter.
There are two important differences between the von Neumann-Fuchs invention and the Teller-Ulam scheme that eventually proved successful. First, in the von Neumann-Fuchs scheme, radiation is used to heat material, not directly to compress it. You could say that von Neumann and Fuchs were considering the energy content of the radiation, while Teller and Ulam were considering its pressure (pressure augmented, even in their scheme, by a hot plasma and evaporation, or ablation, from the imploding container). Second, von Neumann and Fuchs assumed that the “main charge”—the large cylinder of deuterium—would remain essentially uncompressed. Their device would still be a classical Super, just set burning in a new way (a way that, in fact, would not have worked).
Teller, like most other Los Alamos physicists, left the “Hill” in 1945 following the end of World War II. But in 1949, he came back—before the Soviet nuclear explosion in August 1949, before the GAC meeting that October, before the Truman statement of January 31, 1950, before the six-day work week—for what was to be a one-year leave of absence from the University of Chicago, with thermonuclear weapons on his mind. Once the Soviet bomb was detonated and detected, once Truman issued his statement, the stars and planets realigned. Suddenly Teller was not an outlier at the lab. He was in the middle of the lab’s priorities. By early 1950 John Wheeler had joined him, soon to be followed by Lothar Nordheim,{19} John Toll, and me, plus a pair of reassigned young physicists in the lab. Wheeler stayed a year, then headed the ancillary Project Matterhorn at Princeton University for two more. Nordheim stayed for two years, then went back to his academic post at Duke University. In 1951 Teller, piqued at being passed over as head of Los Alamos’s thermonuclear program, returned for a time to the University of Chicago. But he devoted almost the entirety of his career after 1949 to thermonuclear weapons—some of that career back at Los Alamos but most of it at Livermore,[56] a new weapons lab for which he had successfully campaigned.
By the time that H-bomb work shifted into high gear in early 1950, scientists had a pretty clear idea of what the Super might contain and what it might look like, just no idea whether it would work. Within a heavy container, perhaps made of steel and either spherical or cylindrical, would be the thermonuclear fuel, some part of which would, in one way or another, be brought to a temperature of tens of millions of degrees by the explosion of an adjacent fission bomb. The fuel could be liquid—and very cold—deuterium (heavy hydrogen), or deuterium salted with expensive and hard-to-produce tritium (still heavier hydrogen), or a solid substance, lithium hydride, as much as possible of it composed of the isotopes lithium-6 and deuterium (thus often called lithium-6 deuteride). Whatever the fuel, if the device was to work, the temperature would need to be maintained or increased after it was ignited, just as in a fireplace or coal-burning stove. If too much energy was lost, by radiation or in some other way, the thermonuclear flame would not propagate. The device would fizzle.
Just such a fizzle was feared. As I discussed back in Chapter 1, the theorists were concerned that with little or no compression too much energy might be radiated away, leaving behind not enough to keep the flame going.
This was the situation in early 1950. Since computers were then mostly women with desk calculators, and scientists had only their slide rules, it had not yet been possible to carry out calculations with enough precision to reach a definite conclusion about whether any version of the classical Super would work. Some people—including the distinguished mathematician John von Neumann, his wife Klari, the Los Alamos physicists Harris Mayer and Nick Metropolis, and Cerda and Foster Evans (a husband-and-wife team from Los Alamos)—made a herculean effort over the years 1946 to 1950 to wring some answers from what was the world’s first true computer, the ENIAC—built in Philadelphia and then stationed in Aberdeen, Maryland.{20} But that computer, huge of girth and small of brain, was not up to the task, and prospects for the Super remained cloudy. Refined calculations, both with people and with better machines, were evidently necessary.
The people included no less than the esteemed mathematician Stan Ulam and the visiting physicists John Wheeler and Enrico Fermi. The machines included an upgraded ENIAC in Maryland and IBM CPCs (card-programmed calculators) in various locations, with MANIACs in Princeton and in Los Alamos under construction. The fuel of choice for calculations was deuterium with various admixtures of tritium. Lithium-6 deuteride, although more promising in the long term, was harder to analyze. The calculations were challenging enough just with D and DT fuels.
56
Livermore, founded in 1952 as a branch of Berkeley’s Radiation Laboratory, became the separate Lawrence Livermore National Laboratory in 1980.