Tunnel Diode
A tunnel diode is a heavily doped pn junction diode, also called an Esaki diode, whose current in the forward direction first rises, then falls as voltage increases, and then rises again. That falling stretch is a region of negative differential resistance, and it exists because electrons cross the very thin junction by quantum tunnelling instead of climbing over the potential barrier. Discovered by Leo Esaki in 1957 while he worked at Tokyo Tsushin Kogyo, the company that became Sony, the device was the first practical demonstration of tunnelling in a semiconductor and earned Esaki a share of the 1973 Nobel Prize in Physics.
The device works because the doping is so heavy that the junction is only about ten nanometres wide. In an ordinary diode the dopant density is modest and the depletion region is wide, so carriers must gain energy to cross it. In a tunnel diode both sides are doped so heavily, to levels of order 10^19 to 10^20 atoms per cubic centimetre, that the semiconductor is degenerate: the Fermi level lies inside the conduction band on the n side and inside the valence band on the p side, and the depletion layer is thin enough for electrons to pass directly through the barrier to empty states on the other side. Because tunnelling is a quantum effect that needs no thermal activation and little transit time, the diode responds to signals at microwave frequencies.
The current-voltage curve explains its uses. At small forward voltage, tunnelling current rises to a peak, typically a few milliamperes in common parts at a small fraction of a volt. With more voltage the energy bands move out of alignment, the number of states available for tunnelling shrinks, and the current falls to a valley minimum; beyond the valley, the usual diffusion current of a forward-biased diode takes over and rises exponentially. Between peak and valley the slope of current against voltage is negative, so the diode acts like a negative resistance and, with a suitable bias network, can offset the positive resistance in a resonant circuit and keep it oscillating. Ratio of peak to valley current is a key figure of merit, higher for gallium arsenide than germanium, and the device can also be used as a bistable switch or a very fast logic element.
Esaki's discovery began in a study of heavily doped germanium junctions and was reported in 1958. Working on narrow germanium junctions at Tokyo Tsushin Kogyo in 1957, he saw a negative resistance in the forward direction that could not be explained by classical diode theory, and he explained it as tunnelling, building on Zener's earlier theory of tunnelling breakdown in reverse bias. His paper appeared in Physical Review in 1958, and the tunnel diode quickly attracted interest from American laboratories, with several companies producing germanium and gallium arsenide parts in the early 1960s. It was among the first devices in which quantum mechanics could be seen in an electrical measurement, and the 1973 Nobel Prize recognized this, shared with Ivar Giaever for tunnelling in superconductors and Brian Josephson for the Josephson effect.
The tunnel diode was soon displaced, and understanding why prevents a common misunderstanding. It had extremely high speed but low power-handling ability, a small signal swing of a few hundred millivolts, and, being a two-terminal device, no isolation between input and output, which made circuits awkward compared with transistors. Faster, cleaner bipolar and field-effect transistors, Gunn and IMPATT diodes for microwave power, and integrated logic overtook it in the 1960s. The usual misconception is that tunnelling makes carriers cross the barrier faster than light or without any delay; the speed advantage is real but finite and comes from the absence of minority-carrier storage, not from a violation of anything. Another pitfall is expecting it to rectify: in reverse the diode conducts easily, and variants called back diodes use that property.
Its legacy continues in the physics of nanoscale devices. Resonant tunnelling diodes, built from quantum-well heterostructures that Esaki and Raphael Tsu's superlattice ideas helped inspire, still show negative resistance at very high frequencies and are studied for terahertz sources. Tunnelling is now central to flash memory, scanning tunnelling microscopes and tunnel field-effect transistors, and in the opposite direction it is a leakage problem for the thinnest gate oxides, so the quantum effect Esaki saw in germanium shapes modern chip design both as tool and as obstacle.
Read tunnel diode as a junction so thin and so heavily doped that electrons cross it by quantum tunnelling: the resulting negative-resistance dip made it a brilliant, fast microwave device in 1957 and, once transistors improved, a mostly historical one.