A junction that emits light
An electronic component can turn electrical energy into light. At a semiconductor p–n junction, a p-type region, where holes are the majority carriers, meets an n-type region, where electrons dominate. Under forward bias, injected electrons and holes can recombine and release photons. The material determines which transitions occur and how much of the light escapes.
In 1962, Nick Holonyak Jr. and S. F. Bevacqua reported visible emission from gallium arsenide phosphide (GaAsP) junctions. Their paper, received on 17 October and published on 1 December, concerns an alloy composition chosen to favor direct optical transitions. This does not mean that every GaAsP composition has the same band structure.
Red emission and laser action are different results
| Observation in the original paper | Conditions and scope |
|---|---|
| Red light visible to the eye | Devices operated in air at 300 K, with currents from 20 mA to over 100 mA; the authors reported no overheating |
| Laser action near 710 nm | A device at 77 K, using 1–5 µs current pulses; threshold current density about 11,000 A/cm² |
| Spectral narrowing in the same device at 77 K | About 125 Å near or below threshold, then 20 Å at 16,000 A/cm² and 12 Å at 19,000 A/cm² |
The ambient-temperature observation must not be combined with the cryogenic laser conditions to suggest a room-temperature laser. Nor do the reported currents establish efficiency, calibrated brightness or service life.
A light-emitting diode (LED) can produce spontaneous emission without a laser cavity. Laser operation additionally requires optical gain and feedback. The experimental pieces had two polished, parallel opposing faces forming an optical cavity, with the junction perpendicular to those faces. Above threshold, stimulated emission produced a narrower spectrum. The different vertical scales in the paper’s spectra prevent a quantitative gain comparison from peak heights alone.
Why this milestone matters
Alloy composition offered an engineering route to visible semiconductor emitters. Such devices later became indicators, displays and optical components through further improvements in materials, manufacturing, extraction of light and reliability. The 1962 experiment did not establish the performance of those later products.
It was not the first report of electroluminescence: H. J. Round described light from silicon carbide in 1907, and Oleg Losev investigated related emission in the 1920s and 1930s. Efficient blue LEDs required later work; the practical blue-LED developments of 1989–1993 were a separate milestone that helped make white solid-state lighting possible. Red emission in 1962 was not a white lamp.
Methods, evidence and limits
The original two-page article and its figures were read. Holonyak’s 1987 account provides a participant’s retrospective chronology, not an independent replication. The Nobel scientific background and IEEE historical documentation place the result alongside earlier electroluminescence and later blue LEDs.The paper describes semiconductor preparation, junction formation, polished cavity faces, pulsed excitation and spectral measurements. These require specialized fabrication and optical instrumentation. This article is an explanation of the historical record, not a laboratory or home experiment protocol. No semiconductor fabrication, measurement or independent replication was performed for it; raw digital data and executable analysis code were not available in the consulted historical paper.
A small calculation checks the scale: 7,100 Å equals 710 nm. Using E = h c / λ, where h is Planck’s constant, c the speed of light and λ the vacuum wavelength, gives approximately 1.75 eV (electronvolts) per photon. This calculation was executed with the exact International System of Units (SI) constants and λ = 710 nm; it does not reproduce the experiment or measure its efficiency. The wavelength and spectral widths remain the authors’ reported observations.
The cover is an AI-generated conceptual illustration of junction emission, not a photograph, measurement or exact reconstruction of the 1962 device. AI assisted the writing and seven translations; automated and model reviews do not constitute human scientific review.