In April of 1946 about 30 scientists attended a top-secret, three-day conference at Los Alamos, to review the work that had been completed at the weapons laboratory on the hydrogen bomb during World War II. Edward Teller, Klaus Fuchs, Stanislaw Ulam, and John von Neumann were among those to attend.
The first day, Teller described his proposed design, which came to be known as the "classical super." Although details remain classified, Teller's basic concept was that an atomic bomb, might produce enough heat to ignite deuterium (an isotope of hydrogen). Teller proposed putting a fission bomb at one end of a long pipe full of liquid deuterium. According to the theory, the atom bomb would heat one end of the pipe sufficiently to start a burning wave in the deuterium that would proceed down the tube.
At the conference, mathematicians Nicholas Metropolis and Anthony Turkevich explained the results of some very crude calculations they'd done on the thermonuclear reaction earlier on the year. They conceded that in order to make the calculations manageable, they had simplified them which had the effect of producing more ideal conditions than would really exist. But given these conditions, the mathematicians reported that the superbomb described by Teller could release substantial amounts of energy.
Following the conference, Teller and his group drafted a report. It noted that unlike the fission bomb, a thermonuclear would not require a critical mass and so the report concluded the super's "scale is limited only by the amount of deuterium fuel provided. Thermonuclear explosions can be foreseen which are not to be compared with the effects of the fission bomb , so much as to natural events like the eruption of Krakatoa."
The report also noted that calculations done by hand confirmed, "that the system would ignite." But the report conceded that it was not clear whether once ignited the thermonuclear burning would continue." The report concluded that, "the detailed design submitted to the conference was judged on the whole workable." Several scientists who had attended the meeting complained when they saw the first draft of the report, saying that it was far too optimistic. Some suggested revisions, but when the report was issued in June, it was essentially unchanged.
When hand and computer calculations were finally completed on the classical super in 1950, they showed that not only would thermonuclear burning not be sustained in the classical super, but that the system would never ignite a thermonuclear reaction in the first place
Fuchs left Los Alamos less than two months after the conference. But before returning to England, he and von Neumann filed a patent that proposed using radiation to ignite the super. It was an idea that Teller returned to several years later when his initial plan for the H-bomb turned out not to work.