Calvin Fuller

Calvin Fuller (1902–1994) was a Bell Labs chemist who showed how to build p-n junctions in silicon and germanium by diffusing impurity atoms into the crystal, a method that became the backbone of transistor fabrication and, with Gerald Pearson, produced the silicon solar cell announced in April 1954. He is the standard example of a chemist whose work shaped an industry usually pictured as the province of physicists: his contribution was a controllable way of changing the electrical type of a semiconductor from the surface inward, rather than a theory of how it behaves.

Fuller's central idea was that dopants do not have to be present in the melt, because they can be driven into a finished crystal from its surface. Early junction transistors were made by changing the doping while the crystal grew. Fuller studied the alternative: exposing a heated slice to a vapour or source of donor or acceptor atoms, which then migrate inward at a rate that depends on temperature and time. By choosing the impurity, the temperature and the duration, a thin layer of the opposite type could be formed at a depth that could be reproduced from slice to slice. In the 1950s he and his colleagues published measurements of how fast different impurities diffuse in silicon and germanium, the data that process engineers needed to design diffused junctions rather than to guess at them.

With Gerald Pearson, Fuller applied the technique to the solar cell that Daryl Chapin was developing. The cell announced on 25 April 1954 needed a junction close to the illuminated surface, so that light-generated electrons and holes reached the junction field before they recombined. A diffused layer provided exactly that, and the resulting silicon cell was several times more efficient than the selenium devices Chapin had been testing. Fuller's role was the materials chemistry: producing silicon of the right type and forming the junction, while Pearson contributed the physics and measurement of junction behaviour. The three men are credited together, and the pairing of a chemist and a physicist is characteristic of how Bell Labs worked on semiconductors in this period.

The same diffusion methods drove the shift from grown-junction and alloy transistors to diffused ones. A diffused base can be made very thin, much thinner than was practical in a grown crystal, and a thin base means a short transit time for carriers and so a higher operating frequency. Diffusion also works across a whole slice at once and from one side, which suits batch production. Together with the oxide masking of Frosch and Derick in 1957 and the photolithography developed soon after, diffusion became one of the three pillars of the planar process that Hoerni introduced at Fairchild in 1959. Fuller did not invent those later steps, but they inherit the idea he helped to establish: define the device by controlled impurity profiles built inward from the surface.

It is easy to misread his name as that of an inventor of the transistor or the solar cell, when the accurate description is narrower and still important. He did not discover the transistor effect, and the solar cell was a three-person result. His distinct achievement was making diffusion a quantitative engineering tool. A related pitfall is to treat diffusion as a single method: modern fabs use furnace diffusion, ion implantation and other techniques, with implantation taking over precise shallow doping, although thermal diffusion of dopants during heating steps still shapes every profile.

Read Calvin Fuller as the chemist who taught the semiconductor industry to dope a crystal from the outside in: a method of growing junctions by diffusion that powered the first practical solar cell, enabled diffused transistors, and remains a basic idea in every device made today.