The other means became apparent between February and April 1951, following breakthroughs achieved at Los Alamos. One breakthrough was the recognition that the burning of thermonuclear fuel would be more efficient if a high density were achieved throughout the fuel prior to raising its temperature, rather than the classical Super approach of just raising the temperature in one area and then relying on the propagation of thermonuclear reactions to heat the remaining fuel. A second breakthrough was the recognition that these conditions—high compression and high temperature throughout the fuel—could be achieved by containing and converting the radiation from an exploding fission weapon and then using this energy to compress a separate component containing the thermonuclear fuel.

The major figures in these breakthroughs were Ulam and Teller. In December 1950 Ulam had proposed a new fission weapon design, using the mechanical shock of an ordinary fission bomb to compress to a very high density a second fissile core. (This two-stage fission device was conceived entirely independently of the thermonuclear program, its aim being to use fissionable materials more economically.) Early in 1951, Ulam went to see Teller and proposed that the two-stage approach be used to compress and ignite a thermonuclear secondary. Teller suggested radiation implosion, rather than mechanical shock, as the mechanism for compressing the thermonuclear fuel in the second stage. On March 9, 1951, Teller and Ulam presented a report containing both alternatives, titled “On Heterocatalytic Detonations I: Hydrodynamic Lenses and Radiation Mirrors.” A second report, dated April 4, by Teller, included some extensive calculations by Frederic de Hoffmann and elaborated on how a thermonuclear bomb could be constructed. The two-stage radiation implosion design proposed by these reports, which led to the modern concept of thermonuclear weapons, became known as the Teller-Ulam configuration.

The weapons are tested

It was immediately clear to all scientists concerned that these new ideas—achieving a high density in the thermonuclear fuel by compression using a fission primary—provided for the first time a firm basis for a fusion weapon. Without hesitation, Los Alamos adopted the new program. Gordon Dean, chairman of the AEC, convened a meeting at the Institute for Advanced Study in Princeton, New Jersey, hosted by Oppenheimer, on June 16–17, 1951, where the new idea was discussed. In attendance were the GAC members, AEC commissioners, and key scientists and consultants from Los Alamos and Princeton. The participants were unanimously in favor of active and rapid pursuit of the Teller-Ulam principle.

Just prior to the conference, on May 8 at Enewetak atoll in the western Pacific, a test explosion named George had successfully used a fission bomb to ignite a small quantity of deuterium and tritium. The original purpose of George had been to confirm the burning of these thermonuclear fuels (about which there had never been any doubt), but with the new conceptual understanding contributed by Teller and Ulam, the test provided the bonus of successfully demonstrating radiation implosion.

In September 1951, Los Alamos proposed a test of the Teller-Ulam concept for November 1952. Richard L. Garwin, a 23-year-old University of Chicago postgraduate student of Enrico Fermi’s, who was at Los Alamos in the summer of 1951, was primarily responsible for transforming Teller and Ulam’s theoretical ideas into a workable engineering design for the device used in the Mike test. The device weighed 82 tons, in part because of cryogenic (low-temperature) refrigeration equipment necessary to keep the deuterium in liquid form. It was successfully detonated during Operation Ivy, on November 1, 1952, at Enewetak. The explosion achieved a yield of 10.4 megatons (million tons), 500 times larger than the Nagasaki bomb, and it produced a crater 1,900 meters (6,240 feet) in diameter and 50 meters (164 feet) deep.

Further refinements

With the Teller-Ulam configuration proved, deliverable thermonuclear weapons were designed and initially tested during Operation Castle in 1954. The first test of the series, conducted on March 1, 1954, was called Bravo. It used solid lithium deuteride rather than liquid deuterium and produced a yield of 15 megatons, 1,000 times as large as the Hiroshima bomb. Here the principal thermonuclear reaction was the fusion of deuterium and tritium. The tritium was produced in the weapon itself by neutron bombardment of the lithium-6 isotope in the course of the fusion reaction. Using lithium deuteride instead of liquid deuterium eliminated the need for cumbersome cryogenic equipment.

With the completion of Castle, the feasibility of lightweight, solid-fuel thermonuclear weapons was proved. Vast quantities of tritium would not be needed after all. Refinements of the basic two-stage Teller-Ulam configuration resulted in thermonuclear weapons with a wide variety of characteristics and applications. Some high-yield deliverable weapons incorporated additional thermonuclear fuel (lithium deuteride) and fissionable material (uranium-235 and uranium-238) in a third stage. The largest American bombs had yields of 10 to 25 megatons and weighed up to 20 tons. Beginning in the early 1960s, however, the United States built a variety of smaller, lighter weapons that exhibited steadily improving yield-to-weight and yield-to-volume ratios. By the time nuclear testing ended in 1992, the United States had conducted 1,030 tests of weapons of every conceivable shape, size, and purpose. After 1992, computers and nonnuclear tests were used to validate the safety and reliability of America’s nuclear stockpile—though the view was widely held that entirely new computer-generated weapon designs could not be considered reliable without actual testing.