They changed what they were looking for
In 1932 at the Cavendish Laboratory in Cambridge, John Cockcroft and Ernest Walton had built a machine that sent accelerated protons into lithium. An earlier attempt had searched for gamma radiation and found no decisive nuclear signal. When they returned to the problem, they looked instead for heavy charged particles leaving the target. A zinc-sulphide screen gave a small flash when one arrived. Walton saw the characteristic scintillations at a proton energy corresponding to only 125 kilovolts.
The apparatus brought a nuclear experiment under a new kind of control. Natural radioactive sources supplied particles with the energies nature happened to provide. A laboratory accelerator allowed experimenters to change the incident energy and choose a target. To make that useful required an ion source, a high voltage that could be maintained, a tube that would not break down electrically, and a vacuum in which the protons could reach the lithium.
Walton’s engineering work was central. A voltage-multiplying arrangement built high direct voltages from successive charging stages. The machine did not need to fill a large circular hall: its importance lay in producing a controllable proton beam energetic enough to enter a nuclear reaction. The 1951 Nobel prize recognized this pioneering transmutation work, almost two decades after the 1932 observation.
The window helped identify the particles
A flash alone would not say what had happened in the target. Protons might simply scatter and hit the screen. The experiment used a mica window thick enough to stop the incident protons, including those deflected toward the observation side. Particles that emerged through it had a much greater range. Absorbers placed in their path helped measure that range, which could then be related to their energy.
The products behaved like alpha particles, the nuclei of helium atoms. Cockcroft later recalled a simple two-observer coincidence test: two screens and two people showed that particles could emerge in pairs. Better measurements connected the pair with the breakup of lithium-7 after it absorbed a proton. The intermediate system became two helium-4 nuclei. Counting protons and neutrons before and after makes the identity easier to follow: the four protons and four neutrons of the combined initial system end up shared between the two helium nuclei.
The products carried far more kinetic energy than the small incoming proton beam energy. Cockcroft’s historical lecture quoted about 8.6 million electron volts per alpha particle, about 17.2 million in total. Those were the experimental values reported in his account, with the accuracy available at the time. The extra energy came from the lower mass and stronger binding of the final nuclear arrangement. The accelerator started the reaction, and the change in nuclear binding supplied additional energy to the products.
The pair also made the energy accounting visible. The incident proton brought relatively little momentum compared with the two fast outgoing particles. The alpha particles therefore emerged roughly in opposition, with small corrections from the moving initial system. Observing correlated products was more informative than counting an undifferentiated increase in radiation. It connected a specific target nucleus to a specific final arrangement of the same nucleons.
Getting through a barrier did not require climbing over it
The low incident energy had a theoretical reason. A proton is electrically repelled by a positive lithium nucleus. A purely classical particle with too little energy turns back before reaching the region where the strong nuclear force can act. Quantum mechanics instead describes a particle by a wave. A wave can extend through a finite barrier and give a nonzero probability for the particle to appear on the other side. This is tunneling.
Gamow, and independently Gurney and Condon, had used tunneling to explain the escape of alpha particles from nuclei. Cockcroft discussed the reverse problem with Gamow: how readily could an approaching proton enter a light nucleus? The estimate suggested that the achievable laboratory voltages could be enough. The observation of alpha particles then connected the accelerator, the target reaction and a quantum prediction in one experiment.
The method did not make every nuclear reaction equally accessible. The electric barrier grows with the target charge, and the available reaction channels depend on the nucleus and the incident energy. But the lithium result showed that one could study the nucleus by building a beam, changing its energy and reading the products. High voltage had become a tool for asking a specific nuclear question, and the answer appeared as two fast helium nuclei leaving a chosen target.