Semiconductors

Certain crystals embarrassed nineteenth-century electricity. Metals conducted. Glass and rubber insulated. Then a handful of materials started behaving like neither camp: selenium changed conductivity in light, and metal sulfides let current pass more easily one way than the other. Semiconductors entered history not as a triumphant invention but as a category error that laboratory instruments could no longer ignore.

Britain supplied one half of the puzzle in 1873 when Willoughby Smith found that selenium changed its resistance under illumination while testing materials for submarine telegraph cables. Germany supplied the other half a year later when Karl Ferdinand Braun, working in Wurzburg, discovered rectification in metal sulfides. Those results did not look like one coherent field yet. They looked like odd failures of ordinary electrical intuition. Yet together they marked the emergence of semiconductors as usable electrical matter: solids whose conductivity could be shaped by light, contact geometry, temperature, and impurities rather than treated as fixed.

That emergence depended on older tools and concepts. Selenium had to be isolated and characterized first. Sensitive instruments such as the galvanometer had to detect small current changes that earlier experimenters would have missed. The periodic-table gave chemists and physicists a language for treating elements and compounds as members of related families rather than as unrelated curiosities. Even so, nobody in the 1870s possessed quantum mechanics, band theory, or modern crystal-growth methods. Researchers could observe semiconductor behavior long before they could explain it. That gap is part of the story, because semiconductors spent decades as a useful mystery before they became a disciplined platform.

Convergent evolution kept pushing the mystery toward application. Smith reached the class through photoconductivity in the United Kingdom. Braun reached it through asymmetric conduction in Germany. Jagadish Chandra Bose in Calcutta then turned crystal rectification into a practical high-frequency detector during the 1890s, helping make the crystal-detector one of the first major semiconductor devices. Once radio engineers learned that a tiny galena contact could do work that bulkier detectors handled poorly, semiconductors stopped being laboratory oddities. They became components.

That component status created the next layer of niche construction. Demand for better detectors, rectifiers, and later radar parts rewarded purification, crystal growth, and controlled doping. Germanium entered as an especially workable material because mid-twentieth-century labs could purify it before they could do the same at scale with silicon. That is why the first transistor at Bell Labs in 1947 used germanium rather than silicon. Path dependence followed: early semiconductor electronics learned to live inside the manufacturing habits, thermal limits, and contact techniques that germanium made easiest.

Silicon eventually broke that lock-in, not because germanium had failed to demonstrate the principle, but because silicon's oxide and thermal durability suited mass production better. Once engineers mastered purification and the p-n junction, semiconductors underwent adaptive radiation. The same material logic branched into the transistor, the light-emitting-diode, and solar devices that finally gave the old photovoltaic-effect an industrial body. A class of awkward crystals had become a family of devices that could switch, sense, emit, and harvest energy.

Seen across a century, semiconductors look less like a single invention than like a delayed recognition of what certain materials could do. Britain noticed light-sensitive conduction. Germany noticed one-way conduction. India turned that behavior into the crystal-detector. American industrial labs used germanium to build the transistor, and silicon later carried the platform into global scale. Once that chain matured, electronics no longer depended on heating vacuum tubes or moving mechanical contacts. It depended on matter that could be tuned from within.