The Estonian-Irish astronomer Ernst Öpik proposed in 1951 that in the contracting cores of evolved stars (the stars themselves expand to become red giants), the temperature could reach a few hundred million degrees. At these temperatures, Öpik argued, most of the helium would fuse into carbon. Since, however, Öpik’s paper was published in the relatively little-known Proceedings of the Royal Irish Academy, not many astrophysicists knew about it.

Astrophysicist Edwin Salpeter, then at the beginning of his career at Cornell University, didn’t know about it either. In the summer of 1951, Salpeter was invited to visit the Kellogg Radiation Laboratory at Caltech, where the ebullient nuclear astrophysicist Willy Fowler and his group were becoming deeply involved in the study of nuclear reactions thought to be important for astrophysics. Starting with the same idea as Öpik, Salpeter examined the triple alpha process in the hot inferno at the centers of red giants—precisely the problem abandoned by Hoyle’s graduate student. Salpeter immediately recognized that three helium nuclei could hardly be expected to collide simultaneously. It was more likely that two of them might stick together long enough to be struck by a third. Salpeter soon found that carbon could perhaps be produced via a low-probability, two-step process. In the first step, two alpha particles could combine to form a highly unstable isotope of beryllium (⁸Be), and in the second, the beryllium could capture a third alpha particle to form carbon. But there was still a serious problem. Experiments had shown that this particular isotope of beryllium disintegrates back into two alpha particles, with a fleeting mean lifetime of only about 10^-16 seconds (0.00 . . . 1 at the sixteenth decimal place). The question was whether at a temperature of over one hundred million Kelvin, the reaction rate could become so high that some of these ephemeral beryllium nuclei could fuse with the third helium nucleus before falling apart.

When he read Salpeter’s paper, Hoyle’s first reaction was anger with himself for having let such an important calculation slip through his fingers because of the mishap with the graduate student. Upon a closer examination of the entire nuclear reactions network, however, Hoyle estimated that under Salpeter’s assumptions, all the carbon would be transformed into oxygen essentially as fast as it was produced, by fusing with yet another helium nucleus. Some thirty years later, he described this important realization: “Bad luck for poor old Ed, I thought to myself.” (Ed Salpeter was, in fact, nine years younger than Hoyle.) But did this spell disaster for the entire scheme? These were precisely the types of situations in which Hoyle revealed his incredible physical intuition and the clarity of his thought. He started with the obvious: “There has to be some way of synthesizing ¹²C.” After all, not only was carbon relatively abundant in the universe, but carbon was also crucial for life. After evaluating all the potential reactions in his head, Hoyle concluded: “Nothing was better than 3α.” So how could the carbon be prevented from slipping away into oxygen? In Hoyle’s mind, there was only one way: “3α had to go a lot faster than it had been calculated to do [emphasis added].” In other words, beryllium and helium had to be able to fuse together so easily and so quickly that carbon would be produced at a much faster rate than it was destroyed. But what could substantially speed up the rate of carbon synthesis? Nuclear physicists knew of one thing: a “resonant state” in the carbon nucleus. Resonant states are values of the energy at which the probability for a reaction reaches a peak. Hoyle realized that if the carbon nucleus happened to have an energy level that perfectly matched the energy equivalent of the combined masses of the beryllium nucleus and an alpha particle (plus their kinetic energy of motion), then the rate for the fusion of beryllium with an alpha would increase significantly. That is, the probability for the unstable beryllium nucleus to absorb another helium nucleus (alpha particle) to form carbon would be enhanced greatly. But Hoyle did more than merely point out that a resonance would help. He calculated precisely the necessary energy level in the carbon nucleus to obtain the desired effect. Nuclear physicists measure energies in nuclei in units called MeV (an MeV is one million electron volts). Hoyle calculated that for carbon production to match the observed cosmic abundance, a resonant state in ¹²C was needed, at about 7.68 MeV above the lowest energy level (the ground state) of the carbon nucleus. Furthermore, using the known symmetry of the ⁸Be and ⁴He nuclei, he predicted the quantum mechanical properties of this resonant state.

This was all very impressive, except for one “small” problem: No such state was known to exist! The mere idea that Hoyle would be using general astrophysical evidence to make an extremely precise prediction in nuclear physics (much more precise, in fact, than could be calculated based on nuclear physics) was nothing short of preposterous, but Hoyle never lacked chutzpah.

The time was January 1953, and Hoyle was spending a sabbatical of a few months at Caltech. Armed with his new prediction for an unknown energy level of the carbon nucleus, Hoyle went straight into Willy Fowler’s office at Kellogg Laboratory to see whether Fowler and his group could run experiments to verify the prediction. What happened at that meeting has become legendary. Fowler recalled, “Here was this funny little man who thought that we should stop all this important work that we were doing otherwise and look for this state, and we kind of gave him the brushoff. Get away from us, young fellow, you bother us.”