Field-Effect Concept
The field-effect concept is the idea that the current in a conductor can be controlled by an electric field applied across it, from a nearby electrode, instead of by injecting more current into the device. A conducting sheet or channel carries current between two ends; a third electrode, insulated from it, pushes or pulls charge carriers by field alone, widening or narrowing the conducting path. It is a principle about control, distinct from the field-effect transistor entry, which describes the devices that finally realised it.
The appeal of the idea was obvious from the start: a device controlled by voltage that draws almost no control power. The vacuum triode already worked this way, with a grid whose voltage steers a current of electrons through a gap. A solid-state analogue would be small, rugged and without a heated filament. The conducting channel's resistance would be modulated by a field, in the manner of a variable resistor, so a small signal on the control electrode produces a larger change in the main current, which is amplification. The physics seemed straightforward: a field at the surface of a semiconductor should add or remove carriers there, and the sheet resistance should change in proportion.
Early attempts came from several directions, and they were separated by decades from success. Julius Lilienfeld filed patents in Canada in 1925 and in the United States in 1926 describing a layered device with a thin semiconductor, copper sulfide, and a control electrode. Oskar Heil, a German physicist, described a similar field-controlled arrangement in a British patent of 1934. Neither left a record of a working amplifier that could be reproduced. In 1948, William Shockley and Gerald Pearson at Bell Labs built a thin semiconductor film with a nearby metal plate to test the effect directly; theory predicted a sizeable change in conductance, but the measured modulation was far smaller than expected.
Surface states were the obstacle, and understanding them redirected the history of electronics. John Bardeen proposed in 1947 that electrons at the surface of a semiconductor sit in localised states, caused by dangling bonds and imperfections, that trap charge. An applied field then draws electrons into these traps and is screened before it can reach the bulk of the material, so the channel barely responds. The failure of the Shockley experiments pushed Bardeen and Walter Brattain to study the surface instead, and in December 1947 their investigation produced transistor action, in the point-contact device, rather than the field-effect device Shockley had wanted. The principle was set aside for a decade while bipolar transistors, whose action occurs in the bulk, developed.
The way back to the idea required better surfaces, not a better theory. Shockley published a theory of a unipolar field-effect transistor in 1952, avoiding the surface by placing the control gate as a reverse-biased pn junction inside the crystal. The direct, insulated-gate form of the idea had to wait until the late 1950s, when Bell Labs chemists and engineers learned to grow a thermal silicon dioxide on silicon with very few interface traps. With that oxide, a field applied across it could finally reach the silicon, and the transistor of the original concept, now called the MOSFET, became possible. The lesson, a recurring one in physics, is that a correct principle can remain useless until materials science catches up.
A common misconception is that the field-effect idea was a failed rival to the bipolar transistor or that it was simply invented in 1925. A patent describes a hoped-for mechanism; a working, reproducible device that shows gain is a separate achievement, and in this case it took more than thirty years. Another pitfall is thinking the field does the moving: the field only changes how many carriers are available in the channel, while the current itself still comes from the supply voltage. Modern relevance is total: every logic gate, memory cell and AI accelerator relies on a gate field controlling a channel, and the same concept underlies image sensors, touchscreen controllers and chemical sensors that detect ions by their surface fields.
Read the field-effect concept as a control principle that was right in 1925 and impossible until about 1960: a transverse field steering a channel, held up for decades by the quality of a surface.